Summarizing  a  Journal  Article  from  an  Outline   In  the  handout  “Outlining  a  Journal  Article,”  you  learned  how  to  read  a  journal  article,  then  derive  an   outline  from  it.    I  outlined  the  Ceci  and  Bronfenbrenner  article  for  you,  as  a  model.    Now,  I  will  use  that   outline  to  write  a  summary.    First,  recall  my  original  outline:   Outline  of  Ceci  and  Bronfenbrenner  (1985)  Prospective  Memory  in  Children   I. II. III. IV. V. Research  Question(s)/Purpose  of  the  article   The  main  purpose  of  this  study  is  to  examine  a  strategy  called  “strategic  time  monitoring,”   which  is  used  by  children  when  they  have  to  perform  a  prospective  memory  task.     Hypothesis/Hypotheses   Clock  checking  would  be  higher  in  the  laboratory  than  at  home.     Variables   a. Independent  Variable   Location  of  experiment     b. Dependent  Variable   Clock  checking     Methodology   a. Who  participated?   96  children,  half  of  whom  were  10  years  old,  and  half  who  were  14  years  old.     b. What  did  they  do?   They  were  asked  to  bake  cupcakes,  and  put  them  in  the  oven  by  a  certain  time  and  to   remove  them  30  minutes  later.    While  waiting  for  the  cupcakes,  the  children  were  told   they  could  play  PacMan  in  an  adjoining  room.    The  experimenters  also  pointed  out  a   wall  clock.     Major  Findings   The  hypothesis  was  supported.    Children  tended  to  do  more  clock-­‐checking  while  baking   cupcakes  in  the  laboratory  as  compared  to  in  their  home,  which  could  meant  that  children   were  more  comfortable  in  their  home  than  in  the  laboratory.  The  different  colors  have  been   added  to  the  outline  so  you  can  see  where  each  part  of  our  outline  appears  in  our  summary.           The  areas  in  standard,  black  font  are  standard  phrases  that  would  appear  in  a  summary.      So,  here  is   my  summary  (the  font  was  enlarged  so  it  would  be  easier  for  you  to  see  the  colors  –  normally,  you   would  use  12-­‐point  font,  and  the  text  should  be  in  all  black,  and  not  colored):   Ceci  and  Bronfenbrenner  (1985)  conducted  a  study  on   prospective  memory  in  children.    The  main  purpose  of  this   study  is  to  examine  a  strategy  called  “strategic  time   monitoring,”  which  is  used  by  children  when  they  have  to   perform  a  prospective  memory  task.    The  authors  hypothesized   that  clock  checking  would  be  higher  in  the  laboratory  than  at   home.   To  examine  clock  checking,  Ceci  and  Bronfenbrenner   (1985)  varied  the  location  of  the  experiment.    The  study   included  96  children,  half  of  whom  were  10  years  old,  and  half   who  were  14  years  old.    They  were  asked  to  bake  cupcakes,  and   put  them  in  the  oven  by  a  certain  time  and  to  remove  them  30   minutes  later.    While  waiting  for  the  cupcakes,  the  children   were  told  they  could  play  PacMan  in  an  adjoining  room.    The   experimenters  also  pointed  out  a  wall  clock.   Ceci  and  Bronfenbrenner  (1985)  found  that  the  hypothesis   was  supported.    Children  tended  to  do  more  clock-­‐checking   while  baking  cupcakes  in  the  laboratory  as  compared  to  in  their   home,  which  could  meant  that  children  were  more  comfortable   in  their  home  than  in  the  laboratory.  The  different  colors  have   been  added  to  the  outline  so  you  can  see  where  each  part  of   our  outline  appears  in  our  summary.       "Don't Forget to Take the Cupcakes out of the Oven": Prospective Memory, Strategic Time-Monitoring, and Context Stephen J. Ceci and Urie Bronfenbrenner Cornell University CECI, STEPHEN J., and BRONFENBRENNER, URIE. "Don't Forget to Take the Cupcakes out of the Oven": Prospective Memory, Strategic Time-Monitoring, and Context. CHILD DEVELOPMENT, 1985, 56, 152-164. Strategies employed by children in tasks requiring prospective memory (i.e., remembering to do something in the Future) were investigated to illuminate the nature oF the processes involved, and their developmental and contextual determinants. Efficient strategy use was expected to increase with age but to decrease in an unfamiliar setting or on a task associated with higher sexrole expectations. Children were instructed to perform Future activities after waiting 30 min. Children's clock-checking during the waiting period was assessed in their own homes or a university psychophysics laboratory. As predicted, strategic time-monitoring occurred less Frequently in the laboratory than in the home. The anticipated differences associated with higher sex-role expectations and age were most salient For older boys in the laboratory setting. The findings indicate that, when observed in a Familiar setting, children can be shown to employ sophisticated cognitive strategies. Emphasis is placed on the scientific power oFthe laboratory as a contrasting context For illuminating developmental processes. Until recendy, researchers interested in memory have accorded attention almost exclusively to retrospective memory, that is, recalling information about the past. Neglected has been another function of memory important in daily life: remembering to attend to a future event, or prospective memory (Meacham & Leiman, 1982; Meacham & Singer, 1977). ombo, 1980; Wejlman, Ritter, & Flavell, 1975). For example, in their interview study, Kreutzer et al. (1975) asked children how they would go about remembering to bring their ice skates to school the following day. Children frequently reported that they would use an extemal retrieval cue to remind them, for example, placing the skates by the front door or pinning a note to their clothes. Numerous instances of "everyday remembering" attest to the importance of prospective memory, for example, remembering to take vitamins, catch the morning bus, send a spouse an anniversary card, meet with a student, or tum off the bath water in 10 min to prevent a "flood." Yet, despite the ubiquitousness of the phenomenon in our daily lives, there have been few attempts by developmentalists to study strategies used to support prospective remembering, and their developmental and contextual determinants. In this study, we identify and investigate yet another strategy employed by children when called upon to perform a task at some future time. We call it strategic timemonitoring because its use is associated with less frequent clock-checking during a waiting period, thus releasing time for other activities. In addition to providing evidence for the use of this strategy, we examine its development as a function of the age and sex of the child, and investigate ecological factors that interact with these personal characteristics to infiuence efficient cognitive functioning. Investigations of prospective memory have, to date, focused mainly on one aspect: Three distinct phases of strategic timethe use of extemal retrieval cues (Kreutzer, monitoring can be identified: (1) an early "calLeonard, & Flavell, 1975; Meacham & Col- ibration" phase wherein subjects engage in This research was supported by grants from the Cornell University Graduate School and the College oF Human Ecology. Beth Ambinder, Dolores Bradley, and Valencia Norman assisted with the data collection. The authors acknowledge with gratitude the comments on an earlier version oF this paper provided by John A. Meacham, Patricia Worden, and Ulric Neisser as well as the many helpFul suggestions oF three anonymous reviewers and the editor. Address reprint requests to either Stephen J. Geci or Urie Bronfenbrenner, Department oF Human Development and Family Studies, Comell University, Martha Van Rensselaer Hall, Ithaca, NY 14853. [Child Development, 1985, 56, 152-164. © 1985 by the Society for Research in Child Development, Inc. All rights reserved. 0009-3920/85/5601-0013$01.(X)] Ceci and Bronfenbrenner 153 frequent clock-checking to "synchronize their psychological clocks" with the passage of real clock time, (2) an intermediate phase of reduced clock-checking while pursuing other activities, and (3) a "scalloping phase" wherein clock-checking sharply increases as the deadline approaches. Operationally, the use of the strategic time-monitoring strategy is manifested by a U-shaped distribution of such behavior over the course of the waiting period. Similar considerations gave rise to two corollary hypothesbs. First, if children in the laboratory do not epchibit much strategic timemonitoring, what j should be the pattern of their clock-checkiijig during the waiting period? Creater anxiety tends to be accompanied by greater vigilance. Vigilance would be refiected in a constantly rising rate of clock-checking throughout the waiting period. We refer to this phenomenon as anxious time-monitoring. Accordingly: The recent work of Harris and Wilkins (1982), with adults, is consistent with the division of time-monitoring into the three phases just described. The U-shaped distribution of tiieir subjects' clock-checks over time can be interpreted as an initial flurry of clockchecking activity in order to calibrate one's psychological clock, followed by a prolonged period of reduced clock-checking, and finally a relatively rapid burst of "last minute" clockchecks (i.e., a scalloping effect). Hypothesis 2a: Anxious time-monitoring, as manifested by a constantly increasing pace of clock-checking during the waiting period, should be more pronounced in the laboratory setting than in the home. It remains to be demonstrated, however, tliat such a U-shaped distribution of clockwatching over time is indeed efficient, that it does decrease the total amount of monitoring behavior without reducing punctuality, thus releasing time for other activities. Hence, Hypothesis 2b: The total frequency of clock-checking should be greater in the laboratory than in the home. Hypothesis 1: In tasks requiring prospective memory, children who employ a pattem of strategic monitoring, manifested in a U-shaped distribution of clockchecking over time, will spend less total time in clock-watching but still be punctual. As with all behavior, a child's use of strategic time-monitoring can be expected to v:iry as a function of the situation. In particular, we anticipated that children would be more likely to exhibit this more efficient pattem of behavior in the home than in the laboraitory. This expectation was based on Bronfenbrenner's conclusion, drawn from a comparison of studies conducted in both settings (Rronfenbrenner, 1979), that the laboratory, as a typically strange and unfamiliar environment, tends to evoke higher levels of anxiety, especially in children, thereby leading to reduced efficiency in cognitive functioning. Our second hypothesis therefore reads as follows: Second, the combined effect of reduced efficiency and greater strain should result in higher overall levels of clock-checking. Hence, The remaining three hypotheses focus more sharply on developmental issues relating to the age and sex of the child and the manner in which these personal characteristics interact both with the nature of the task and the setting in which it is performed. Hypothesis 3: Since the use of more advanced cognitive strategies is expected to increase with age, resort to strategic time-monitoring will be greater among older children, with a corresponding reduction of total time spent in clock-watching. By contrast, younger children will make less use of strategic monitoring, and spend more time in clock-watching, particularly in the unfamiliar laboratory setting. What should be the effect of age on anxious time-monitoring? On the one hand, older children, being more mature, might be expected to feel less insecure, especially in an unfamiliar situation. On the other hand, being older, they may feel a greater responsibility to accomplish the assigned task. Because of these conflicting considerations, we could not make a specific a priori prediction on this score. Hypothesis 2: Because the pre cesses of socialization enResort to strategic time-monitoring, as manifested courage different e ipectations for boys than in a U-shaped distribution of clock-checking, will be more pronounced when tasks requiring prospec- girls (Maccoby & facklin, 1974; Shepherdtive memory are presented in the familiar environ- Look, 1981), we ai lticipated that children's ment of the home as opposed to an unfamiliar lab- time-monitorin;ig would vary systematically as oratory setting. a joint function oft! e sex of the child and the 154 Child Development nature of the task. Our specific hypothesis was based on the following consideration. Tasks associated with higher expectations for a particular sex are likely to engender more anxiety than activities not regarded as sexappropriate, with a corresponding reduction in efficiency. Our fourth hypothesis expresses this prediction in more specific form: Hypothesis 4: On a traditionally female sex-typed task, such as baking, girls will make less use of strategic timemonitoring. Instead, they will exhihit a more anxious pace of clockwatching than boys and spend more time in the process. Correspondingly, on a traditional male sex-typed task, such as charging a motorcycle battery, boys will exhibit the analogous pattem. Because of increasing sex-role expectations with age, these relationships should be more pronounced with older than with younger children. Finally, on a more general plane, we assumed that the familiarity of the setting would affect not only the overall level of a particular pattem of clock-checking, but also the extent to which this pattern would be affected by other factors relating to characteristics of the child or the situation. To state our final hypothesis in more specific form: Hypothesis 5: The joint influence of age, sex, and nature of the task on patterns of time-monitoring will be greater in the laboratory setting than in the home. Like Hypothesis 2, this prediction derives from Bronfenbrenner's (1979) comparative analysis of studies conducted in the laboratory versus naturalistic settings. One of the conclusions suggested in the review was that, in the less familiar setting of the laboratory, research findings were more likely to be influenced by external factors such as social class, age, and sex differences, or systematic variations in experimental procedure. Method Subjects.—Ninety-six children, 48 10year-olds (mean age = 10.7 years, SD 9 months) and 48 14-year-olds (mean age = 14.6 years, SD 11 months) were recruited through their college-aged siblings and offered $5 either to bake cupcakes or to charge a motorcycle battery. Half of the children at each age were girls. Occupational information was available for 67 of the 96 children's parents. Based on the Hodge-Siegel-Rossi Occupational Prestige Equivalents (Davis, Smith, & Stephenson, 1980), there were no significant group differences in socioeconomic status as a function of age, sex, context, or task assignment, nor was there any amonggroup variation in the percentage of singleparent households or of homes where the mother was employed. Procedure.—All children were tested individually. Those children asked to bake cupcakes were instructed to place them in the oven by a specified time and to remove them 30 min later (as per Jiffy Mix instructions). While waiting, the children were invited to make unlimited use of a Pac Man video machine, beginning 15 min prior to placing the cupcakes into the oven and continuing during the entire 30-min baking period. (The 15-min warm-up period was provided to familiarize children with the game.) At the start of the 30min baking period, children's attention was directed to a wall clock, and they were instructed to remove the cupcakes 30 min from the time shown, for example, "It's now o'clock. Be sure to let them bake exactly 30 min. So, you should remember to take them out of the oven at exacdy o'clock." (Conveniendy, no children were wearing wristwatches on the day of their participation.) The Pac Man game was always placed in an adjoining room sufficiently far from the oven to preclude the use of various extemal retrieval cues such as oven buzzers or the aroma of the maturing cupcakes. The game was bonded to a table top with suction cups. This was done so that the child's back would bave to face the wall clock when playing Pac Man. Thus time-monitoring behaviors (turning in one's seat to check the wall clock) could easily be recorded by an unobtrusive observer who was in the room, pretending to be reading a book or magazine. For children assigned to the motorcycle battery charging task, the same verbal instructions were employed as for the baking task, except that references to baking were replaced with references to battery charging. The children were instructed to remove the battery charger cables after the battery had been charged 30 min. They too were provided with the unlimited use of Pac Man beginning 15 min before the task began and continuing for the entire 30-min charging period. There were two contexts for each task, with 24 children of each age and sex (six per cell) assigned at random to each setting. Half of the children did the baking or charging in their own homes, the other half in a psychophysics laboratory. In the home, Pac Man was set up in whichever room children Ceci and Bronfenbrenner normally played games, provided there was a clock on the wall and a table on which the game could be affixed. All of the children in both settings in fact did play Pac Man while waiting to remove either cupcakes or the battery-charging cable. In the home, the children's older siblings, who were undergraduates trained by the experimenters, acted as unobtrusive observers during the waiting period. While pretending to be reading in another part of tbe room, they recorded each instance when their younger sibling turned away from the game in order tp check the time. None of the observers was informed about the purpose and design of the experiment, or the reasons for recording clock-checking, until after the study had been completed. Pilot testing of the recording procedure indicated nearly perfect reliability among a group offiveof the observers who subsequently participated in the experiment. In the laboratory context, children were brought to a university psychophysics laboratory by their older brothers or sisters. The baking or battery-charging instmctions were given by an unfamiliar age-mate of their older sibling, who was also trained to record the children's clock-watching while they played Fac Man in the laboratory. The child's older sibling was not present during the experiment in the laboratory. Thus, the two contexts differed both in the physical characteristics of die setting (home fumishings vs. psychophysics instrumentation) and in the familiarity of tlie persons present (older sibling vs. unfamili£ir adult). Debriefing.—Children were debriefed by their older siblings. They were informed that the primary purpose oftibestudy had not been to bake cupcakes (or to recharge a battery) but to gain insight into how they kept track of time. It was explained that it was sometimes necessary for researchers to mislead subjects in order to gain knowledge, but ttiat deception was never an ethically attractive way to deal with other people and was used with some reluctance by the researchers. The older siblings explained that there was no right or wrong way to monitor the passage of time. Children were encouraged to ask any questions they had conceming their participation. Finally, they were thanked for their participation and given the agreed-upon $5 stipend. 155 Results Pilot studies and preliminary analyses.— As a precaution, I we sought to determine whether there were any systematic age or sex differences in children's interest and/or proficiency in playing Pac Man, inasmuch as the presence of such differences could affect . the capacity of the game to occupy boys' and girls' attention equally. At the outset of the study, children had been asked to indicate how frequently they played Pac Man (daily, biweekly, weekly, bimonthly, or less often) and how much they enjoyed it {not at all to very much). In addition, gains achieved between initial and final scores, as well as the number of games they played during the waiting period, were recorded as measures of children's concentration on the game. Each of these measures was entered in a multivariate linear regression as a predictor of clockchecking for each 5-min interval. In no instance did any of these variables account for significant variation in children's clockchecking. Somewhat surprisingly, girls expressed slightly more interest in Pac Man than boys and claimed to play it equally often. For the baking task only, all of the older siblings happened to be females. Because of the possibility of confounding by sex of older sibling, another preliminary ansJysis was undertaken. Since, in the battery-charging task, half of the older sil^lings were males, the sex of the older sibling!was used as a predictor of clock-checking fre({[uency in a discriminate function. The resultant Wilks's lambdas indicated that, at least on this task, the sex of the older sibling was nbt associated with variation in the frequency df clock-checking by boys and girls. Principal findings.—With the foregoing possibilities of confounding eliminated from serious consideration, we proceed to an examination of results beiaring on the main hypotheses of the study. For this purpose, children's frequencies of clock-checking were analyzed by means of a five-way analysis-of-variance design (Age x Sex x Context x Task x Time Interval), with the first four factors being between-subjects dichotomies and the final factor being a repeated measure within subjects. In order tc test for the hypothesized U-shaped distribution of clock-checking, six successive 5-min intervals were analyzed for both linear and quadratic components.^ The method of orthogonal polynomials was em- All between-subject main effects and interaction efFects were tested using 1 and 80 df and MSe = .69. All within-subject effects were tested using 5 and 400 df and MSe = .49. Linear and quadratic analyses employed 1 and 80 df and MSe's oF 1.48 and 1.10,respectively. Neuman-Keuls tests were significant at p < .01 unless stated otherwise. 156 Child Development 14-Year Olds 5 - UI I U g I II I I L 10-Year Olds u o 2 17.0 S 14.5 < 3 LU 2 10 15 20 25 HOME CONTEXT 30 10 15 20 25 30 LAB CONTEXT FIG. 1.—Children's clock-checking Frequencies on the cupcake-baking task, late responders excluded mean total number oF clock-checks). ployed for this purpose since it permits testing for the presence of a specifically parabolic function rather than a more general quadratic curve (Fisher, 1929; Winer, 1971). A test of goodness-of-fit revealed that the residuals from linear and quadratic regression were small and nonsignificant. (The linear and quadratic components together accounted for 91% of the within-subjects variation associated with the six 5-min time intervals.) Hence, these two regression components can be regarded as accurate summary measures of the distribution of clock-checking over time. To tum to the findings themselves. Figures 1 and 2 depict the distribution of clockchecking during the waiting period for each of the 16 subgroups of subjects included in the experimental design. Data points along each graph show average frequencies of clock-checking for successive 5-min intervals, with the mean totals for the 30-min periods recorded at the right of each curve. Hypothesis 1 The first hypothesis requires that strategic monitoring, as manifested by a Ushaped distribution of clock-checking, be associated with less total time looking at the clock. Evidence in support of this hypothesis is initially apparent from an inspection of Figures 1 and 2. Perhaps the most striking feature of the two figures is the contrast in the shapes of the curves appearing in the left and right halves of each diagram. With but one exception (to be discussed below), all of the parabolas appear at the left, whereas the curves at the right are generally linear. Hence, an approximate test of the first hypothesis can be made by comparing overall time-checking scores for each successive pair of setting contrasts. As indicated by the means cited in each panel, in every paired comparison, the total amount of clockchecking was greater in the laboratory setting than in the home. This pattem of consistent contrasts is reflected in a highly significant Ceci and Bronfenbrenner 157 U-Year Olds 5 - 5 10 15 20 25 HOME CONTEXT 10 15 20 25 30 LAB CONTEXT FIG. 2.—Children's clock-checking frequencies on the battery-charging task, late responders excluded main effect for Context, F = 59.8, p < .001. In substantive terms, children observed in the home not only made greater use of strategic time monitoring, as indicated by the larger quadratic components, but they also spent less clock-watching time in the process. child in clock-checking and the degree of his or her anxious time-monitoring as measured by the corresponding linear regression coefficient. If our theoretical expectations are correct, the resulting correlation should again be substantial, but this time positive in sign. The obtained coefficient is in fact .67, indicating, not surprisingly (though not necessarily), that children who exhibited a pattem of anxious time-monitoring engaged in more clockchecking overall. = mean total number oF clock-checks). A more precise measure and test of the first hypothesis is provided by examining this same relationship at the level of individual subjects, that is, calculating a correlation between the £unount of time spent by each child in clock-checking, on the one hand, and, on the other, a measure of the extent to which the child resorted to strategic monitoring. Such a measure is obtainable by csdculating, for each child, the quadratic regression coefficient for the distribution of his or her own clock-watching over the 30-min waiting period. The resulting correlation for the total sample was, as expected, rather high and negative in sign, r = — .78, p < .01. It is instructive to calculate the analogous conelation between total time spent by each Taken together, these correlational results also constitute evidence of construct validation for our interpretation of the two principal pattems of clock-checking that have been distinguished: strategic and anxious time-monitoring. Hypothesis 2 The consistently contrasting pattems in Figures 1 and 2 provide even stronger support for the second hypothesis and its two corollaries. Strategic monitoring was indeed more pronounced in the more familiar setting 158 Child Development of the home, whereas anxious time-monitoring was more evident in the laboratory and resulted in more time spent in total clockwatching. This fact is confirmed statistically by highly significant interactions of Context both by linear, F = 54.8, p < .001, and by quadratic Time Intervals, F = 30.7, p < .001. Hypothesis 3 With respect to the effects of Age, we find, as anticipated, that adolescents spent less time in clock-watching than did 10-yearolds. Their respective means for the half-hour waiting period were 11.2 and 14.1, F = 24.2, p < .001. On the average, older children also made more use of strategic time-monitoring, as revealed by a significant Age X Time quadratic, F = 4.8, p < .05. Further analysis of quadratic trends, however, revealed that the age difference was significant only in the laboratory setting, F = 30.7, p < .001. Tbe same qualification applies to the infiuence of Age on anxious time-monitoring, but the effect is in the opposite direction. While using a more efficient strategy of clock-checking than younger children, adolescents of both sexes nevertheless showed a stronger tendency to increase their clock-checking over time. This finding is consistent with the interpretation that older children felt a greater pressure to perform the assigned task on time. Taken together, the contrasting signs of these interaction effects also demonstrate that the two strategies of time-monitoring are not simply opposite sides of the same coin, since they are not always correlated in the same direction. Hypothesis 4 The effects of Age on both types of clockwatching were further qualified simultaneously by the Sex of the child and nature of the Task, F = 5.8, p < .01, for the Age x Sex X Context X Task quadratic trend. To summarize in substantive terms, older boys, again only within the laboratory setting, tended to be most efficient and least anxious when asked to bake cupcakes. The above finding clearly speaks to the fourth hypothesis of the study: the prediction that strategic monitoring will be reduced, and the frequency and pace of clock-checking increased, on a task associated with higher sex-role expectations for the group in question. There was only partial support, however, for the remainder of the hypothesis: the corresponding relationship for girls, while in the expected direction, was not statistically significant, that is, girls did not engage in reliably less clock-checking on the battery-charging task than their female peers on the baking task. It was further anticipated that tbese sexrole effects would be more salient in older children than in younger children. The findings cited above are in accord with this expectation with respect to the distribution of clock-watching over time. Older boys, particularly in an unfamiliar laboratory setting, exbibited a more anxious and less efficient clock-checking strategy when requested to disconnect a battery charger (a male-typed task) than when asked to take cupcakes out of the oven (a female-typed task). The results on the overall fi-equency of time-monitoring also provide corroboration for the fourth hypothesis. As specified in a significant Sex x Task interaction, F = 24.2, p < .001, boys looked at the clock more often in the battery-charging task, and girls in the baking task, particularly when the latter was conducted in the home. None of these relationships was qualified by tbe Age of the child, nor were there any significant main effects for Sex or Task. Hypothesis 5 Finally, as the reader may have noted, a familiar refrain resounds throughout the last few paragraphs: the phrase "only in the laboratory." This recurrence refiects the fact that significant interaction effects were, almost without exception, limited to the laboratory setting. In accord with our final hypothesis, this finding supports the view that the laboratory, as an unfamiliar and tbereby somewhat anxiety-arousing environment, is more likely to activate variation in research findings as a function of characteristics of both the person (e.g., age and sex) and the immediate situation (e.g., the nature of the task). Taken as a whole, the results of this study indicate that, especially when observed in a familiar environment, children as young as 10 years of age can employ a fairly sophisticated cognitive strategy for monitoring the passage of time, one that substantially reduces the overall burden of clock-checking. The question may be raised, however, whether the reduction in amount of clock-checking is achieved at some sacrifice to punctuality. Specifically, did children in fact remember to take the cupcakes out of the oven before they were burned, or to disconnect the battery cable before the battery burned out? And if some cbildren forgot, was the failure in prospective memory related to the strategy of clock-watching that they employed? When prospective memory fails.—There were in fact 14 children who were more than Ceci and Bronfenbrenner 159 Cupcake baking task Battery charging task a 10-YEAR OLDS O U 4 O 3 14 YEAR OLDS 3 a S5.5 5 10 15 20 25 30 5 10 15 20 25 30 FIG. 3.—Clock-checking frequencies of "late responders" in the HOME context. No cases oF late responding occurred in the LAB context (E = mean total number oF clock-checks). 90 sec late in removing cupcakes from the oven or removing the battery-charger cable, and an additional seven wbo were between 60 and 90 sec late. Seven instances of late responding involved 10-year-oIds on tbe baking task, six of whom were boys. All seven of diese children had been assigned to the home context. Another eight instances of late responding involved 10-year-olds on the battery-charging task, five boys and three girls. All but one had been assigned to the home context. Six older children (all but one of whom were boys) responded late, three on die baking task and three on the charging tiisk. All had been assigned to tbe home context. Thus, there was only a single instance of "forgetting to remember" in the laboratory context. And, overall, boys were more tban three times as likely to forget as were girls (16 vs. 5). Since nearly all of the forgetting occurred in the home and the home was also the context for most of the strategic time-monitoring, thie question arises whether the latter type of behavior is indeed effective. Does it not occasionally result in failure to perform the task on time? To check on this possibility, we examined the pattern of clock-checking for the late responders as a group. As is apparent from Figure 3, these late Eirrivals did not engage in strategic monitoring. This observation is supported statistically by the high point biserial (r = — .81, p < .01) between the regression coefficients fbr tbe quadratic components and lateness versus "on time." This result confirms that, to a marked extent, those who responded late did not employ strategic timemonitoring. An analysis of individual late responders' pattems of clock-checking revealed that the magnitude of their quadratic coefficients did not exceed chance in a single instance (all p's > .10). Moreover, an inspection of Figure 3 reveals that late responders approached the task differendy from their "on-time" peers. First, as previously noted, they neither engaged in strategic monitoring, nor did they increase their clock-watching over time. Instead, to the extent that they showed any variation in rate, it was to decrease the frequency of checking as they approached^ the deadline. Even more striking is the low overall level of clockchecking that characterized this group, a total of less than six checks for the entire 30-min waiting period conqpared with 10—16 checks for the sample as a whole (see Figs. 1 and 2). Given this weak apd waning pattern of response, it is not surprising that they forgot to perform the task on' time. 160 Child Development The deviant behavior of the late responders also suggests a more general inference. A reduction in the overall level of clockchecking can apparently result from any of several processes. One of these is the use of a strategic pattem of monitoring; another may reflect a low level of effort accorded to the monitoring task. Moreover, these processes can lead to somewhat different consequences. On the one hand, the fact that none of the late responders had tised a calibration strategy suggests that such a strategy insures against missing the deadline while at the same time reducing the total time expended in monitoring. On the other hand, reducing clockwatching for other reasons can result in "forgetting to remember." For example, an analysis of the relationship between late responding and the absolute frequency of clockcbecking revealed that the children who checked the clock least often were the latest in removing cupcakes or battery cables, Q = .81, p < .01, by Yule's Coefficient of Association. Tbis result replicates a similar finding in an earlier study by Harris and Wilkins (1982). The foregoing analyses of data from late responders permit an extension of our earlier generalization. Strategic monitoring not only reduces the overall burden of clock-watching, but also accomplishes this economy without risk to the effective functioning of prospective memory; one does not "forget to remember." Discussion Tbe principal conclusions of this experiment are most appropriately presented in the context of the specific hypotheses of the study. Thus we believe our findings clearly demonstrate that children as young as 10 years of age are able to employ fairly sophisticated cognitive strategies in support of prospective memory. Whether they in fact do so, however, depends in significant degree on the context in which prospective memory is activated. Our findings show equally clearly that children are far more likely to use a complex, time-conserving strategy in the familiar environment of the home than in the unfamiliar setting of a psychophysics laboratory. By contrast, the latter milieu evokes a pattem of response involving an escalating pace of clock-cbecking suggestive both of greater caution and greater anxiety. There is no evidence, however, that either of these strategies is more effective than the other in terms of achieving the ultimate goal of remembering to carry out a specified task at some future time. Those children who failed to make the deadline employed neither of the above pattems of clock-checking but rather telegraphed their failure by reducing the frequency with which they were verifying the passage of time. There are indications, however, that the two effective strategies have rather different trade-offs. Moreover, these are trade-offs that children 10 years of age can already recognize and act upon. Thus it would appear that what we have called anxious time-monitoring insures against missing the deadline but does so at tbe sacrifice of intervening time that could be employed for otber purposes. Strategic time-monitoring, by contrast, offers an intervening period for otber activities, but possibly at some risk of failing to perform the required task on time. Our findings suggest that children perceive the request to carry out a future task as a less demanding, and perhaps less iniportant, responsibility when tbe task is to be performed in the presumably more relaxed atmosphere of the home than in the less farhiliar and possibly more prestigious environment of the laboratory. Correspondingly, children in the latter setting, even though already capable of using a less time-consuming strategy, take precautions to insure that they will not fail to live up to expectations. It is a nice instance of Freud's principle of "anxiety in the service of the ego." The interpretation of our results as a function of increased tension evoked by the strange and unfamiliar setting of a psychophysics laboratory requires furtber comment. Although not without sOme theoretical and empirical basis, tbis explanation is nevertheless an inference not grounded in any data from the experiment itself. In order to provide such direct data, we carried out an additional subexperiment making possible a further and more critical test of our a priori explanation. A test case: A halfway house between laboratory and home.—For this purpose, an additional group of 24 children were asked to perform the cupcake-baking task in a standard kitcben located in a former residential apartment of a home economics building. While waiting, children played Pac Man in an adjacent living room. In all other respects the procedure was identical with that followed in the laboratory setting. Thus we were now introducing a third context that, in terms of familiarity, was intended to be intermediate between the other two: at one extreme, one's own kitcben; tben an ordinary kitchen in an unfamiliar but compatible environment (i.e., the home economics building); and, at the other extreme, the psychophysics laboratory. Tbe results for this intermediate context are quite straightforward, and readily sum- Ceci and Bronfenbrenner marized as follows. For the younger children, the pattem of findings in the home economics kitchen was practically indistinguishable from that for the laboratory. Thus there was little evidence of strategic monitoring, the pace of clock-checking increased with time, and the overall firequency was only slightly lower than that for the laboratory (14.5 and 15.7, respectively). It was substantially and significantly higher, however, than the corresponding figure for the home (12.3). In sharp contrast, the results for the 14-year-olds in the home economics kitchen were scarcely distinguishable from those for the home; both of tiiese groups showed high levels of strategic monitoring, and correspondingly low frequencies of overall clock-watching (10.2 for tlie home economics setting and 9.5 for the home setting vs. 13.5 for the laboratory). The foregoing findings are consistent with a hypothesis of an increase in strategic monitoring and a decrease in the frequency and pace of clock-checking as a joint function of the age of the child and the familiarity of the setting. For a 10-year-oId, an unfamiliar kitchen in a home economics building at a university is no less strange than a psychophysics laboratory and has similar effects on reducing the efficiency of cognitive strategies that support remembering. But for an adolescent, an ordinary kitchen in one location is no less familiar than in another, whether the place is a home or a university building, with the result that cognitive strategies remain efficient and unaffected. The results of the "kitchen-on-campus" experiment also shed some light on which aspects of context evoke differences and similarities in pattems of time-monitoring. Following the usual practice in a laboratory experiment, we employed as observers in that setting research assistants who were unknown to the subjects of the study. A similar procedure was followed in the "kitchen-oncampus" experiment. Hence any differences in styles of clock-checking in these two contexts cannot be attributed to the presence of a familiar versus unfamiliar observer. Accordingly, the more anxious and less efficient patte:m of time-monitoring exhibited by older children in tbe laboratory setting must have been induced by some features peculiar to that environment that also had special significance for youngsters of high school age. Such features might include, for example, the profusion of elaborate technical equipment or, at a more abstract level, the prestigefiil and perhaps mystifying aura of science. With respect to younger children, however, the infiuence of interpersonal relation- 161 ships cannot be ruled out since the contrast in their pattern of nionitoring in home versus out-of-home environments was confounded by the presence of an older sibling in the home setting. Hence, for tbis age group, tbe possible effect of a family member's presence on the cognitive strategies employed by the child remains an issue for future research. The laboratory and ecological validity.— It is clear that, had our experiment been conducted only in the laboratory, we would have reached rather different conclusions. In particular, we would have underestimated many children's ability to employ a fairly sophisticated cognitive strategy in order to "remember not to forget." Recall that in the laboratory setting there was little evidence for strategic monitoring. Instead, children exhibited what we now recognize as comparatively high levels of clock-watching. Moreover, with the exception of older boys in the baking task, the frequency of clock-watching increased steadily during the course of the half-hour waiting period. Over a decade ago, developmentalists voiced concern as to whether principles developed from laboratory experiments would generalize to "everyday settings" (Brown, Bransford, Ferrara, & Campione, 1983; Stevenson, 1970; White, 1970). While these arguments were not based on empirical comparisons, otbers bave recently reported substantial differences in the effects of what is ostensibly the same behavior observed in the lab and nonlab settings (Acredolo, 1979; Bickman & Henchley, 1972; Cole, Hood, & McDermott, 1982; DeLoache, 1980; Graves & Glick, 1978; Luria, 1982; Strayer, 1980). In perhaps the seminal investigation of the lack of correspondence between laboratory and naturalistic measures of remembering, Istomina (1975) showed that children recalled approximately twice as many items from a word list when it was presented as a shopping list as opposed to a laboratory study of memory. This and other demonstrations of the noncorrespondence between naturalistic memory and laboratory remembering (e.g.. Cole et al., 1982) have prompted some researchers to conderrin laboratory settings, and even tbe analytical Apparatus tbat experimentalists bring to th^ task of gathering new knowledge (e.g.. Cole et al., 1982; Gergen 1978). In our view, the results of the present study are more pro^jerly construed not as a criticism of laborato^ procedures per se, but rather as a demonstration of the limitations involved in relying exclusively on a laboratory context for clues to children's competen- 162 Child Development cies on everyday tasks. Our findings point to the as yet unexploited power of the experimental laboratory as an ecological contrast tbat belps to highlight the distinctive features of other settings as contexts of development (Bronfenbrenner, 1977, 1979), including tbe development of cognitive strategies. "Calibrating one's psychological clock": Fact or fiction?—It is possible to argue that the high levels of early clock-checking did not refiect children's use of the calibration strategy. After all, the deployment of the calibration strategy was only an inference. We shall contend, however, that the type of timemonitoring characteristic of high levels of early clock-checking must have been qualitatively different from the type of timemonitoring employed by children who engaged in low levels of early clock-checking. In short, more than the numbers of clockcbecks were different between these groups. Otherwise, how could children with high levels of early clock-checking have been able to reduce their subsequent checking well below the levels exhibited by those with low levels of early clock-checking? It seems that the most parsimonious explanation for this tumabout in clock-checking behavior is to consider that those children who began with high levels of clock-checking were comparing the passage of real time with their subjective judgments of the length of time that had gone by; tbat is, they were calibrating their psycbological clocks. It would appear tbat acquired confidence in one's estimation of time, derived through an early comparison process, was what permitted children with high levels of early clock-checking to reduce such bebavior substantially during what they estimated to be the intervening period of waiting. Caveat lector.—Three shortcomings of the study need to be acknowledged. First, tbe present design did not permit an assessment of which aspects of the various contexts were implicated in the observed results. Further work is required to determine the importance of factors such as familiar furnishings, the presence or absence of older siblings, a "scientific" atmosphere, or some combination of these or other features in producing the observed effects. Second, one may legitimately wonder how specific the observed findings were to: (a) the use of a 30-min waiting period, (fo) the particular tasks, (c) the instmctions provided, and {d) the use of Pac Man as the intervening task. Third, some might claim that what we are calling timemonitoring is not a memory function but rather a process of attention-directing. We agree that time-monitoring is a form of attention-directing, but contend that this process is integrally involved in prospective remembering. Tbus, we have referred to it as "a cognitive strategy that supports prospective memory," much like other support processes (e.g., rehearsal, metamnemonic awareness, knowledge-based strategies, etc.). Tbat the tasks we employed entailed prospective memory can hardly be disputed; 21 out of tbe 96 children who participated in these tasks in fact forgot to fulfill the assigned responsibility. Conclusion Tbe present findings would appear at first glance to run counter to the view that complex forms of memory monitoring ai-e undeveloped in young children (see Flavell & Wellman, 1977). In this study, children as young as 10 exhibited a reasonably welldeveloped type of monitoring during the waiting period. Of course there is no way to be certain of our inference about what the children were doing during the waiting period, short of remnning the experiment using clocks programmed to mn faster or slower than normal—which is being done presendy. Attempts tbrough postexperimental interviews to elicit information about cbildren's awareness of using calibration proved uninformative. To the extent that children employed a calibration strategy, tbey appeared to do so spontaneously, without effort or awareness, for they were unable to confirm its use during interviews. Recendy, research has begun to reveal many forms of processing that are undertaken by children without awareness or effort (see Ceci, 1983, 1984, for reviews). Ceci and Howe (1982) have provided a framework and some empirical support for the position that some forms of monitoring occur " . . . without intention, and neither the process of directing the attention, or the resultant product of this attention, ever became conscious. . . . We have seen that parts of the cognitive system appear to be outside the control of the thermostat (metamemory)" (p. 162). Thus, the present findings raise the interesting specter of a fairly complex strategy being automatically deployed. If confirmed, this would help explain why young children appear to be adept at its use, as automatic processing has been shown to be age-invariant, provided that a minimal level oi experience has occurred (Ceci, 1983). In conclusion, we suggest that this series of experiments has some broader implications for research on development-in-context. First, the developmental significance of strategies Ceci and Bronfenbrenner employed to support retrospective memory is highlighted by tbe results of administering the present experiment in a nonconventional kitchen in a university building. For 10-yearolds, the kitchen may as well have been a university laboratory; for adolescents, it may as well have been their own kitchen at home. For younger children, one room at a university was as strange as another; for the adolescents, a kitchen was a kitchen regardless of where it was located, and it connoted none of the special challenge and concem experienced by young high school students exposed to the world of sophisticated science. Even the latter, however, can lose its stirring power if the task to be performed seems unimportant. In the laboratory setting, adolescent boys steadily escalated their clockchecking lest they fail to disconnect the motorcycle battery cbarger on time. But when the task to be performed was remembering to remove cupcakes from the oven, there was little speedup during the waiting period; instead, the 14-year-old boys, after spending the first few minutes calibrating their psychologica^l clocks, relaxed into a pace of less frequent monitoring until a few minutes before the expected deadline, when they again began to check the clock more often. No such temporary respite was shown by adolescent girls. For them, apparently, remembering to take cupcakes from the oven was no less challenging in a scientific setting than was disconnecting a battery charger. None of the subtle yet psychologically significant differences in responding would have been observed, however, had the experiment been carried out solely in the home. Earlier we commented that the conclusions drawn from the study would have been different had it been conducted only in the laboratory; clearly the converse of this statement is equally true. This reciprocal relationship testifies to the scientific importance and power of the laboratory as a contrasting context for illuminating developmental processes and outcomes. Such a view of the laboratory's potential is hardly new. In a recent paper, Markova (1982) calls attention to George Herbert Mead's (1934) strong stand on this same point: ". . . i t was Mead who vehemently defended the laboratory against those wbo claimed that all psychological phenomena should be explored and could be understood only in terms of individual experience. Arguing against phenomenologists and positivists at the same time, he claimed that it was wrong to overemphasize the artificiality of the experimental apparatus and technique of the 163 psychological laboratory (Mead, 1938, p. 35). They are a necessary part of a psychological investigation because they enforce a specific and exact kind of human conduct which otherwise would not be available for investigation" (Markova, 1982, p. 197). The repetition in the laboratory of experiments or observations conducted in a naturalistic setting like the home, or vice versa, offers rich and as yet unexploited possibilities for revealing the remarkable capabilities of children, from the earliest ages onward, to respond to their environments in ways that are cognitively complex and discriminately adaptive both to the opportunities and the risks that these environments present for the child's development. References Acredolo, L. P. (1979). Laboratory versus home: The effect of environment on the 9-month-oId inFant's choice of spatial reFerence system. Developmental Psychology, 15, 666-667. Bickman, L., & Henchley, T. (1972). Beyond the laboratory: Field research in social psychology. New York: McGraw-Hill. BronFenbrenner, U. (1977). Toward an experimental ecology of human development. American Psychologist, 32, 513-531. BronFenbrenner, U. (1979). The ecology of human development: Experiments by nature and design. Cambridge, MA: Harvard University Press. Brown, A. L., BransFord, J. D., Ferrara, R. A., & Campione, J. C. (1983). Learning, remembering and understanding. In J. H. Flavell & E. M. Markman (Eds.), Handbook of child psychology (Vol. 3, pp. 77-166). New York: Wiley. Ceci, S. J. (1983). Automatic and purposive semantic processing characteristics oF nonnal and language/leaming-disabled children. Developmental Psychology, 19, 427-439. Ceci, S. J. (1984). Leaming disabilities and memory development. Journal of Experimental Child Psychology, 38, 356-376. Ceci, S. J., & Howe, M. J. A. (1982). Metamemory and the efFects oF attending, intending, and intending to attend. In C. Underwood (Ed.), Aspects of consciousness (Vol. 3, pp. 147-164). London: Academic Press. Cole, M., Hood, L., & McDermott, R. (1982). Ecological niche picking: Ecological invalidity as an axiom oF experimental cognitive psychology. In U. Neisser (Ed.), Remembering in natural context (pp. 336-341). San Francisco: W. H. Freeman. Davis, J. A., Smith, T. W., & Stephenson, C. B. (1980). Occupatiorial classification distributions. General social surveys, 1972—1980: Cu- 164 Child Development m.ulative codebook. Chicago: National Opinion Research Center. DeLoache, J. S. (1980). Naturalistic studies oFmemory For object location in very young children. In M. Perlmutter (Ed.), New directions for child development (pp. 147-163). San Francisco: Jossey-Bass. Fisher, R. (1929). The design of experiments. Edinburgh: Oliver & Boyd. Flavell, J. H., & Wellman, H. M. (1977). Metamemory. In R. V. Kail & J. W. Hagen (Eds.), Perspectives on the development of memory (pp. 116-132). HiUsdale, NJ: Erlbaum. Gergen, K. (1978). Experimentation in social psychology: A reappraisal. European Journal of Social Psychology, 8, 507-527. Grave, Z. R., & Glick, J. (1978). The effect of context on mother-child interaction: A progress report. Quarterly Newsletter of the Institute for Comparative Human Development, 2, 4 1 46. Harris, J., & Wilkins, A. (1982). Remembering to do things: A theoretical Framework and an illustrative experiment. Human Learning: Journal of Practical Research Applications, 1, 1-14. Istomina, Z. M. (1975). The development of voluntary memory in pre-school-age children. Soviet Psychology, 13, 5-64. Kreutzer, M. A., Leonard, G., & Flavell, J. H. (1975). An interview study of children's knowledge about memory. Monographs of the Society for Research in Child Development, 40 (1, Serial No. 159). Luria, A. R. (1982). Language and cognition. New York: Wiley Interscience. Maccoby, E., & Jacklin, G. W. (1974). The psychology of sex differences. StanFord, GA: StanFord University Press. Markova, I. (1982). Paradigms, thought and language. New York: Wiley. Meacham, J., & Golombo, J. A. (1980). Extemal retrieval cues Facilitate prospective remembering in children. Journal of Educational Research, 73, 299-301. Meacham, J., & Leiman, B. (1982). Remembering to perForm future actions. In U. Neisser (Ed.), Remembering in natural contexts (pp. 327-335). San Francisco: W. H. Freeman. Meacham, J. A., & Singer, J. (1977). Incentive efFects in prospective remembering. Jourrwl of Psychology, 97, 191-197. Mead, G. H. (1934). Mind, self, and society. Chicago: University oF Chicago Press. Shepherd-Look, D. L. (1981). Sex differentiation and the development oF sex roles. In B. Wolman (Ed.), Handbook of developmental psychology. Englewood CliFFs, NJ: Prentice-Hall. Stevenson, H. W. (1970). Leaming in children. In P. H. Mussen (Ed.), Carmichael's manual of child psychology (Vol. 1, pp. 849-938). New York: Wiley. Strayer, J. (1980). A naturalistic study oF empathic behaviors and their relation to aFFective states and perspective-taking skills in preschool children. Child Development, 51, 815-822. Wellman, H. M., Ritter, K., & Flavell, J. H. (1975). Deliberate memory behavior in the delayed reactions oF very young children. Developmental Psychology, 11, 780-787. White, S. H. (1970). The leaming theory tradition For child psychology. In P. H. Mussen (Ed.), Carmichael's manual of child psychology (Vol. 1, pp. 657-702). New York: Wiley. Winer, B. J. (1971). Statistical principles in experimental design (2d ed.). New York: McGrawHill. Emotion 2009, Vol. 9, No. 2, 282–286 © 2009 American Psychological Association 1528-3542/09/$12.00 DOI: 10.1037/a0014904 Viewing Cute Images Increases Behavioral Carefulness Gary D. Sherman, Jonathan Haidt, and James A. Coan University of Virginia Infantile physical morphology—marked by its “cuteness”—is thought to be a potent elicitor of caregiving, yet little is known about how cuteness may shape immediate behavior. To examine the function of cuteness and its role in caregiving, the authors tested whether perceiving cuteness can enhance behavioral carefulness, which would facilitate caring for a small, delicate child. In 2 experiments, viewing very cute images (puppies and kittens)—as opposed to slightly cute images (dogs and cats)—led to superior performance on a subsequent fine-motor dexterity task (the children’s game “Operation”). This suggests that the human sensitivity to those possessing cute features may be an adaptation that facilitates caring for delicate human young. Keywords: cuteness, nurturance, care, fine-motor skill Standard laboratory dexterity tasks score performance as the number of objects successfully moved per second. Because cuteness may not make people faster (only more careful), we used a similar task that was not time dependent: the classic children’s game “Operation” (Hasbro, Pawtucket, RI), in which participants use tweezers to remove small objects (body parts) from confined spaces. This task is similar to standard fine-motor dexterity tasks (e.g., the O’Connor tweezer dexterity task, Lafayette Instrument, Lafayette, IN), but performance can be quantified without reference to speed. Because positive actions directed toward a child likely require physical gentleness, we also used a grip-strength gauge as a measure of physical weakness/gentleness. In addition, during the viewing of the slide show we monitored heart rate and electrodermal responding. This allowed us to (a) detect changes in autonomic physiology that might facilitate finemotor coordination, and (b) to assess whether any shifts in behavior can be attributed to general physiological arousal. Finally, given that responsiveness to cuteness may be rooted in maternal caregiving, and given that women are generally more responsive to cuteness than are men (e.g., they smile more at cute children; Hildebrandt & Fitzgerald, 1978), we tested only women in Experiment 1. In Experiment 2, we tested both men and women. Several factors entered into our choice of images to use as stimuli in these experiments. Some studies of cuteness have used simple schematic drawings as stimuli (e.g., Alley, 1983b). This approach allows for the manipulation of the size and proportion of specific craniofacial features (e.g., eye size), but the stimuli tend to be relatively weak as emotion elicitors. When assessing self-report outcomes (e.g., hypothetical willingness to defend the child), this may not be problematic. Given our interest in manipulating carefulness, however, we believed that more powerful stimuli were necessary. We used photographs of real animals, young and mature. Humans are highly attuned to the physical features that characterize their young, such as a large rounded forehead, large low-set eyes, and a small chin (Alley, 1981, 1983a; Huckstedt, 1965). Those who possess these features are deemed “cute” and are the object of a variety of nurturing and affectionate impulses, such as high-pitched vocalizations (i.e., “baby talk”; Spindler, 1961; Zebrowitz, Brownlow, & Olson, 1992), preferential looking (Hildebrandt & Fitzgerald, 1981), leniency (McCabe, 1988), and protectiveness (Alley, 1983b). This research suggests that the tendency to respond emotionally to infantile physical features may promote the provision of care, especially to infants, who are otherwise helpless due to their physical and neural immaturity. However, the exact ways in which cuteness may enhance care are not yet fully understood. Cuteness might simply and only strengthen adults’ emotional attachments to infants, thereby increasing their willingness to care for them. Alternatively, the affective “cute response” may include a behavioral component that facilitates caregiving itself. Because caring for a small, delicate child requires one to act with great care, we reasoned that cuteness cues might stimulate increased attention to, and control of, motor behavior. We therefore predicted that seeing cuteness will increase behavioral carefulness. In two experiments, we tested this prediction by having participants view a slide show that contained images of animals. We varied the age—and thus cuteness— of the animals experimentally (varying the age of animals depicted in photographs influences perceived cuteness across a range of species, including dogs and cats; Sanefuji, Ohgami, & Hashiya, 2007). Because high levels of carefulness seem more critical for fine-motor movements (e.g., brain surgery) than for gross-motor movements (e.g., running), we used performance on a fine-motor dexterity task as an index of behavioral carefulness. Gary D. Sherman, Jonathan Haidt, and James A. Coan, Department of Psychology, University of Virginia. The research presented in this paper was supported in part by a National Science Foundation Graduate Research Fellowship to Gary D. Sherman. We thank Jesse Graham, Selin Kesebir, and Patrick Seder for their help. Correspondence concerning this article should be addressed to Gary D. Sherman, Department of Psychology, University of Virginia, P.O. Box 400400, Charlottesville, VA 22904-4400. E-mail: gds6d@virginia.edu Experiment 1 Method Participants. Forty University of Virginia undergraduate women participated for partial course credit (mean age ⫽ 18.46). 282 BRIEF REPORTS Participants were randomly assigned to one of two conditions (low cuteness or high cuteness). Stimuli. The slide show consisted of three sections, each with nine images. The first (baseline) and third (post) sections featured neutral images of house interiors and were identical in content across conditions. The middle (main) section featured images of animals— either puppies and kittens (high cuteness) or dogs and cats (low cuteness), depending on condition. Each image was presented for 10 s, with a 3-s black screen providing a transition between images. We validated the main stimuli by having an independent sample (N ⫽ 17) rate the images on cuteness and interestingness using 6-point scales ranging from 0 (not at all) to 5 (extremely). Compared to the low-cuteness images, the high-cuteness images were considered cuter (M ⫽ 3.98 vs. M ⫽ 1.37), t(16) ⫽ 15.21, p ⬍ .001, and more interesting (M ⫽ 2.44 vs. M ⫽ 1.77), t(16) ⫽ 3.35, p ⬍ .01. Measures. A Biopac MP100 System (Biopac Systems, Goleta, CA), sampling at 1,000 Hz, was used for physiological data acquisition. An electrocardiogram (ECG) measured heart activity via three general purpose electrodes attached to each participant (Lead 1 configuration). Acqknowledge 3.7.2 software (Biopac Systems, Goleta, CA) extracted interbeat intervals (IBIs), which were then visually inspected and manually corrected for artifacts. An absolute measurement of skin conductance level (SCL) was obtained by placing two electrodes on the volar surface of the medial phalanges of the first and third fingers of the nondominant hand. We used Redux Electrolyte Paste (Parker Laboratories, Fairfield, NJ) as a conductant. To compute the change in heart rate (HR) and SCL from baseline to the main section, we computed mean HR and SCL levels for these sections using CMet Software (for HR; Allen, 2002; available from http://apsychoserver.psych.arizona.edu) and Acqknowledge 3.7.2 software (for SCL). A slide show after the questionnaire assessed emotional impact (“How much did this slide show affect you emotionally?”), physical impact (“How much did this slide show affect you physically?”), and peak intensity of specific emotions (happiness, entertainment, amusement, calmness, tenderness, sadness, fear, and surprise) on a 6-point scale ranging from 0 (not at all) to 5 (extremely). At the end of the experiment, participants used the same 6-point scale to rate the cuteness and interestingness of the animal slide show. In addition, mood was assessed at the beginning of the experiment and after the slide show using a 100-point scale ranging from 1 (the worst I have ever felt), to 100 (the best I have ever felt). In the operation task, the participant used a pair of tweezers to remove various small plastic body parts from the “patient” without touching the tweezers to the sides of each compartment. Participants had one chance to remove each of 12 body parts, and performance was scored as the number of body parts successfully removed. Finally, grip strength was measured using a hand dynamometer. Procedure. The experimenter told participants (who were run individually) that their physiology would be monitored while they performed several tasks. After a partial hookup was complete (ECG was not attached yet to keep participants’ dominant hand free for the behavioral tasks), participants were given the hand dynamometer and asked to squeeze it as hard as possible. Participants then played the Operation game while the experimenter 283 observed unobtrusively and recorded their scores. Next, the ECG was attached and participants moved to the viewing chamber (a separate area of the experiment room, enclosed by a curtain), were seated, and watched the slide show on a projection screen. Afterward, the ECG was removed. Participants then squeezed the hand dynamometer again, completed the questionnaire, and then played the Operation game once again. Participants then answered the final two self-report items. Results and Discussion The means for all self-report variables are presented in Table 1, along with the results of t tests of the difference in the mean for each item as a function of condition. The most intensely experienced emotions— happiness, calmness, tenderness, amusement, and entertainment—were experienced more intensely in the highcuteness condition than in the low-cuteness condition ( ps ⬍ .05 for all except calmness, for which p ⬍ .10) and the high-cuteness slide show was rated as more interesting and cuter than the low-cuteness slide show ( ps ⬍ .05). The mean change for each behavioral and physiological variable as a function of condition is presented in Table 2. As predicted, cuteness increased performance on a subsequent task requiring extreme carefulness: participants showed significantly greater improvement in performance (from before to after the slide show) on the operation task in the high-cuteness condition than in the low-cuteness condition, t(38) ⫽ 1.99, p ⫽ .05, d ⫽ 0.63. Contrary to predictions, the manipulation of cuteness did not influence change in grip strength (t ⬍ 1). HR did increase more in the high-cuteness condition (M ⫽ 1.64) than in the low-cuteness condition (M ⫽ .02), t(37) ⫽ 1.89, p ⫽ .07, d ⫽ 0.61, but no effect of cuteness on change in SCL was observed (t ⬍ 1). Because the stimuli used in the two conditions differed on a variety of dimensions beyond cuteness (e.g., interestingness, positivity), we examined the correlations between change in operation performance and the self-report variables. Change in performance was positively correlated with ratings of cuteness (r ⫽ .29, p ⫽ .07) and self-reported intensity of tenderness (r ⫽ .34, p ⬍ .05) but not with ratings of interest (r ⫽ .09, p ⫽ .60) or intensity of happiness, amusement, entertainment, calmness, sadness, or surprise (rs between ⫺.07 and .16, ps ⬎ .33). This pattern suggests that the effect of cuteness on participants’ carefulness in executing fine-motor movements was likely due to the images’ cuteness and tenderness-inducing qualities rather than their general positivity or interestingness. Experiment 2 In Experiment 1, viewing images of puppies and kittens enhanced fine-motor performance, supporting the hypothesis that cuteness increases behavioral carefulness. Although the pattern of correlations suggests that cuteness was the critical dimension responsible for this effect, the images differed on several dimensions, precluding us from ruling out other extraneous influences (e.g., positive affect). In Experiment 2, we aimed to replicate the main finding of Experiment 1 (that cuteness increased carefulness) with two entirely new sets of stimuli that were pretested to match across conditions in the levels of positive affect and interest they evoked. This matching allows us to isolate the effect of cuteness BRIEF REPORTS 284 Table 1 Means (SE) for Self-Report Items by Experiment and Condition Experiment 1 Variable General Emotional impact Physical impact Change in mood Emotion Happiness Calmness Tenderness Amusement Entertainment Surprise Sadness Fear Ratings Interesting Cute Experiment 2 Low cute High cute p Low cute High cute p 1.35 (.21) .40 (.15) ⫺.79 (2.21) 2.40 (.23) .65 (.21) 6.29 (1.88) .002 .34 .02 2.14 (.27) .93 (.24) 4.12 (1.49) 1.54 (.23) 1.00 (.19) 2.54 (1.63) .09 .81 .48 1.75 (.34) 2.45 (.36) 1.30 (.31) .85 (.21) .60 (.21) .60 (.22) .50 (.19) .20 (.12) 3.60 (.27) 3.40 (.37) 2.55 (.39) 2.30 (.31) 1.90 (.36) .60 (.27) .55 (.27) .10 (.10) ⬍.001 .07 .02 ⬍.001 .003 1.00 .88 .52 2.82 (.28) 3.00 (.33) 2.04 (.29) 2.14 (.29) 1.39 (.24) .75 (.19) .50 (.19) .46 (.16) 2.71 (.30) 3.57 (.24) 2.54 (.30) 2.29 (.31) 1.29 (.27) .43 (.17) .43 (.14) .21 (.12) .79 .17 .23 .74 .77 .22 .76 .21 1.75 (.25) 3.15 (.27) 2.60 (.31) 4.75 (.10) .04 ⬍.001 2.57 (.27) 3.07 (.31) 2.50 (.27) 4.29 (.20) .85 .002 Note. Ratings were given on 6-point scale, ranging from 0 (not at all) to 5 (extremely). p ⫽ result of t test of difference in means between conditions. from these factors that may alter fine-motor performance independently. We also included male participants to test whether the effect of cuteness on carefulness was specific to women. Method Participants. Fifty-six University of Virginia undergraduates participated for partial course credit (23 women, 33 men; mean age ⫽ 18.80). We tested for, and report, any main or interactive effects of gender. Stimuli. Because the two sets of stimuli used in Experiment 1 differed on more dimensions than cuteness, we created a new low-cuteness image set that included more interesting dog images as well as several images of lions and tigers. Although this expands the range of animals to include more exotic felines (that are rarely kept as pets), this was necessary to achieve a level of interest, emotional power, and positivity similar to that of the puppy and kitten images. In addition, to enhance the generalizability of any findings, we used a new set of puppy and kitten images. An independent sample (N ⫽ 12) assessed the following dimensions: cute, interesting, enjoyable, and exciting using 6-point scales ranging from 0 (not at all) to 5 (extremely). Compared to the lowcuteness images, the high-cuteness images were rated as being cuter (M ⫽ 4.43 vs. M ⫽ 2.86), t(16) ⫽ 9.23, p ⬍ .001. The two sets of images, however, were equally interesting (M ⫽ 3.37 vs. M ⫽ 3.50), equally enjoyable (M ⫽ 3.67 vs. M ⫽ 3.40), and equally exciting (M ⫽ 1.55 vs. M ⫽ 1.89) (all ts ⬍ 1.63, ps ⬎ .12). Measures and Procedure. The measures and procedure of Experiment 2 were identical to those of Experiment 1, except for one minor change: after the slide show, completion of the questionnaire followed playing the Operation game. This change was made to rule out the possibility that the effect of cuteness on carefulness in Experiment 1 was due to the secondary act of reflecting on, and writing about, one’s emotional experience (of tenderness and other emotions) rather than the primary act of viewing the images. Table 2 Mean Difference Scores (SE) for Physiological and Behavioral Variables by Experiment and Condition Experiment 1 Experiment 2 Variable Low cute High cute p Low cute High cute p Change in: Operation Grip strength HR SCL .60 (.44) ⫺3.35 (1.56) .02 (.46) ⫺.79ⴱ (.15) 1.80ⴱ (.41) ⫺4.35 (2.33) 1.64ⴱ (.72) ⫺.67ⴱ (.21) .05 .72 .07 .64 .46 (.29) 1.93 (1.31) .73ⴱ (.35) ⫺.76ⴱ (.18) 1.32ⴱ (.33) 2.43 (1.16) 1.17† (.63) ⫺.59ⴱ (.15) .05 .78 .54 .48 Note. Operation measured in body parts (maximum ⫽ 12). Grip strength was measured in pounds per square inch. Heart rate (HR) was measured in beats per minute. Skin conductance level (SCL) was measured in micromhos. p ⫽ significance level for test of whether difference scores varied by condition (independentsamples t test). Difference scores marked with a symbol differed significantly from zero (one-sample t test). ⴱ p ⬍ .05. † p ⬍ .10. BRIEF REPORTS Having participants wait to report on their emotional experience until after the fine-motor dexterity task allowed our carefulness measure to immediately follow the viewing of the images. Results and Discussion The means for all self-report variables are presented in Table 1, along with the results of t tests of the difference in the mean for each item as a function of condition. The two sets of stimuli used were nearly identical except for the critical dimension of cuteness. Notably, the two conditions did not differ on change in mood or on any specific emotion. This was true even of tenderness—an emotion putatively related to cuteness—although the difference between the means was in the predicted direction.1 This means that the two slide shows were matched on the degree to which they elicited positive affect. Therefore, our cuteness manipulation was effective in isolating the specific effect of cuteness from the more general potential influences of positive emotion or mood. In addition, it is noteworthy that our inclusion of several lion and tiger images as low-cuteness stimuli did not amplify any negative emotions. For example, the self-reported intensity of fear did not differ by condition and was extremely low in both conditions. The mean change of each behavioral and physiological variable as a function of condition is presented in Table 2. Replicating the main finding of Experiment 1, participants showed significantly greater improvement on the operation task in the high-cuteness condition than in the low-cuteness condition, t(54) ⫽ 1.97, p ⫽ .05, d ⫽ 0.48. Although there was a trend for women (d ⫽ 1.03) to show a larger effect of condition than men (d ⫽ 0.24), this was not statistically significant: Gender ⫻ Condition interaction, F(1, 52) ⫽ 1.36, p ⫽ .25. As in Experiment 1, cuteness did not affect change in grip strength (t ⬍ 1). Unlike in Experiment 1, in which cuteness was associated with increased HR, neither change in HR nor change in SCL differed by condition (ts ⬍ 1). Together, this pattern makes it unlikely that the observed effect of cuteness on operation performance was due to general physiological arousal. Although change in HR did not differ by condition, it increased slightly in both conditions, perhaps because viewing pictures of animals (whether high or low in cuteness) triggers excitement and an approach orientation. Unlike in Experiment 1, change in operation performance was not significantly related to any self-report items (rs between ⫺.13 and .19, ps ⬎ .15). The failure to replicate the relationship between self-reported tenderness and behavioral carefulness found in Experiment 1 may have been due to the longer delay in Experiment 2 between the slide show and completion of the questionnaire. The further self-report emotional assessments get from the event in question, the more people’s responses tend to reflect their beliefs about emotion rather than emotion itself (Robinson & Clore, 2002a, 2002b). Thus, self-reports of tenderness in Experiment 2 may have been less indicative of the intensity of experienced tenderness than those obtained in Experiment 1. 285 tion), an effect that cannot be attributed to general positivity (e.g., mood or specific positive emotion) or arousal (measured via selfreport and autonomic physiology). This behavioral shift toward increased carefulness makes sense as an adaptation for caring for small children, and is consistent with the view that cuteness is a releaser of the human caregiving system (Lorenz, 1950/1971). Moreover, this finding suggests that cuteness does not just influence one’s willingness to engage in caregiving behaviors but also influences the ability of one to do so. That is, cuteness not only compels us to care for cute things but also prepares us to do so via its effects on behavioral carefulness. This finding fits nicely with the embodied cognition perspective that emphasizes the way affective states are constrained by, and expressed in, the body (Barrett & Lindquist, 2008). Our finding suggests that the tenderness elicited by something “cute” is more than just a positive affective feeling state—it can literally make people more physically tender in their motor behavior. Research has demonstrated an ideomotor effect whereby the processing of positive stimuli facilitates pulling a lever, the basic motor behavior involved in pulling desired objects closer (presumably reflecting a behavioral predisposition for approach; Chen & Bargh, 1999; Rotteveel & Phaf, 2004). The current finding may be a novel manifestation of the extension of this effect beyond simple valence-approach/avoid relationships. Having a specific kind of positive affective orientation toward an object (finding it cute and experiencing tender feelings) can influence the specific kind of motor actions one is prepared to make (careful, tender movements). Contrary to predictions, cuteness did not make people any weaker, at least as we measured it. It is possible that had we not instructed participants to squeeze as hard as possible (which may have amplified variance associated with trait strength and limited variance associated with state strength), that cuteness would have had a noticeable effect. That is, cuteness may not make people physically weaker but may make them less willing to exert their full strength. Another concern is that grip strength is often used as a measure of motivation. If cuteness triggered an approachoriented motivational state this may have counteracted any shift toward gentleness. This is the first investigation to document that immediate shifts in carefulness—indexed here by fine-motor performance— can be elicited by cuteness cues. This suggests that two factors—the importance of physical contact in early mammalian development and the extremely delicate nature of human young—may have exerted evolutionary pressures favoring those who could respond to the presence of cues colloquially described as “cute” with increased carefulness. 1 Women reported more tenderness and sadness and rated the images as cuter and more interesting, regardless of condition, than did men. References General Discussion In two experiments, we found that exposure to images of young, cute animals (kittens and puppies) increased performance on a task that demanded extreme carefulness in order to successfully execute finely tuned motor movements (the children’s game Opera- Allen, J. J. B. (2002). Calculating metrics of cardiac chronotropy: A pragmatic overview. Psychophysiology, 39, S18. Alley, T. R. (1981). Head shape and the perception of cuteness. Developmental Psychology, 17, 650 – 654. Alley, T. R. (1983a). Age-related changes in body proportions, body size, and perceived cuteness. Perceptual and Motor Skills, 56, 615– 622. 286 BRIEF REPORTS Alley, T. R. (1983b). Infantile head shape as an elicitor of adult protection. Merrill-Palmer Quarterly, 29, 411– 427. Barrett, L. F., & Lindquist, K. A. (2008). The embodiment of emotion. In G. R. Semin & E. R. Smith (Eds.), Embodied grounding: Social, cognitive, affective, and neuroscientific approaches (pp. 237–262). New York: Cambridge University Press. Chen, M., & Bargh, J. A. (1999). Consequences of automatic evaluation: Immediate behavioral predispositions to approach or avoid the stimulus. Personality and Social Psychology Bulletin, 25, 215–224. Hildebrandt, K. A., & Fitzgerald, H. E. (1978). Adults’ responses to infants varying in perceived cuteness. Behavioural Processes, 3, 159 –172. Hildebrandt, K. A., & Fitzgerald, H. E. (1981). Mothers’ responses to infant physical appearance. Infant Mental Health Journal, 2, 56 – 64. Huckstedt, B. (1965). Experimentelle untersuchungen zum “kindchenschema.” [Experimental studies on the “baby schema”]. Zeitschrift fur Experimentelle und Angewandte Psychologie, 12, 421– 450. Lorenz, K. (1971). Studies in animal and human behaviour, Vol. 2. (R. Martin, Trans.). Cambridge, MA: Harvard University Press. (Original work published 1950). McCabe, V. (1988). Facial proportions, perceived age, and caregiving. In T. R. Alley (Ed.), Social and applied aspects of perceiving faces (pp. 89 –95). Hillsdale, NJ: Erlbaum. Robinson, M. D., & Clore, G. L. (2002a). Belief and feeling: Evidence for an accessibility model of emotional self-report. Psychological Bulletin, 128, 934 –960. Robinson, M. D., & Clore, G. L. (2002b). Episodic and semantic knowledge in emotional self-report: Evidence for two judgment processes. Journal of Personality and Social Psychology, 83, 198 –215. Rotteveel, M., & Phaf, R. H. (2004). Automatic affective evaluation does not automatically predispose for arm flexion and extension. Emotion, 4, 156 –172. Sanefuji, W., Ohgami, H., & Hashiya, K. (2007). Development of preference for baby faces across species in humans (Homo sapiens). Journal of Ethology, 25, 249 –254. Spindler, P. (1961). Studien zur vererbung von verhaltensweisen. 3. Verhalten gegenuber jungen katzen. [Studies on the transmission of behavioral patterns. 3. Behavior toward young cats]. Anthropologischer Anzeiger, 25, 60 – 80. Zebrowitz, L. A., Brownlow, S., & Olson, K. (1992). Baby talk to the babyfaced. Journal of Nonverbal Behavior, 16, 143–158. Received July 17, 2008 Revision received October 31, 2008 Accepted November 10, 2008 䡲 E-Mail Notification of Your Latest Issue Online! Would you like to know when the next issue of your favorite APA journal will be available online? This service is now available to you. Sign up at http://notify.apa.org/ and you will be notified by e-mail when issues of interest to you become available!
Purchase answer to see full attachment