The study examined how people represent the number four on a clock face with Roman numerals when drawing from memory versus copying. Subjects either copied a clock face, drew it from memory with warning, or drew it from a surprise memory task. Those who copied were more likely to accurately draw "IIII" for four, while those in memory conditions often drew "IV" instead due to reliance on schemas. In a later recall task, copy subjects who initially drew "IIII" were more likely to incorrectly recall it as "IV" compared to memory condition subjects. The results support the idea that copying relies less on schemas than recalling from memory.
The study examined how people represent the number four on a clock face with Roman numerals when drawing from memory versus copying. Subjects either copied a clock face, drew it from memory with warning, or drew it from a surprise memory task. Those who copied were more likely to accurately draw "IIII" for four, while those in memory conditions often drew "IV" instead due to reliance on schemas. In a later recall task, copy subjects who initially drew "IIII" were more likely to incorrectly recall it as "IV" compared to memory condition subjects. The results support the idea that copying relies less on schemas than recalling from memory.
The study examined how people represent the number four on a clock face with Roman numerals when drawing from memory versus copying. Subjects either copied a clock face, drew it from memory with warning, or drew it from a surprise memory task. Those who copied were more likely to accurately draw "IIII" for four, while those in memory conditions often drew "IV" instead due to reliance on schemas. In a later recall task, copy subjects who initially drew "IIII" were more likely to incorrectly recall it as "IV" compared to memory condition subjects. The results support the idea that copying relies less on schemas than recalling from memory.
The study examined how people represent the number four on a clock face with Roman numerals when drawing from memory versus copying. Subjects either copied a clock face, drew it from memory with warning, or drew it from a surprise memory task. Those who copied were more likely to accurately draw "IIII" for four, while those in memory conditions often drew "IV" instead due to reliance on schemas. In a later recall task, copy subjects who initially drew "IIII" were more likely to incorrectly recall it as "IV" compared to memory condition subjects. The results support the idea that copying relies less on schemas than recalling from memory.
On 1! September zc1c Access details Access Details [subscription number vzec!evcc] Publisher Psychology Press lnforma ltd Registered in lngland and Wales Registered Number 1czv1 Registered office Mortimer House, !- 11 Mortimer Street, london W1T !JH, UK Memory Publication details, including instructions for authors and subscription information http//www.informaworld.com/smpp/title-contentt1!ec!!c Does Clock-watching Make You Clockwise` Anne Richards To cite this Article Richards, Anne(1vve) 'Does Clock-watching Make You Clockwise`', Memory, 1 1, 1v c To link to this Article DOl 1c.1ccc/11v1cee1 URl http//dx.doi.org/1c.1ccc/11v1cee1 Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf This article may be used for research, teaching and private study purposes. Any substantial or systematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply or distribution in any form to anyone is expressly forbidden. The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material. Does Clock-watching Make You Clockwise? Anne Richards Birkbeck College, University of London, UK Christopher C. French Goldsmiths College, University of London, UK Peter Harris University of Hertfordshire, UK French and Richards (1993) found that subjects asked to draw from memory a clock that had Roman numerals on its face typically represented the number four on the clock face as ``IV rather than the correct ``IIII, whereas those merely asked to copy it typically drew ``IIII. The current experiments followed the methodology of French and Richards, but then went on to examine the subsequent memorial representation of the number four. Subjects drew a clock with Roman numerals on its face, either from memory (with or without forewarning) or while the clock remained in full view. Subsequently, subjects were asked to recall the exact form in which the numbers were represented on the clock (Experiment 1) or were asked to recognise which of two clocks had been presented earlier (Experiment 2). Findings supported the idea that subjects in the copy condition were more likely than subjects in other conditions to draw the clock without invoking schematic knowledge of Roman numerals. The basic effect reported by French and Richards was replicated in both experiments. Furthermore, in both experiments, subjects who correctly drew the clock with the number four represented as ``IIII were more likely to misrepresent it as ``IV in the subsequent memory task if they were in the copy condition rather than the two memory conditions. The results are interpreted in terms of schema theory. I N T R O D U C T I O N Clocks with Roman numerals typically represent the number four as ``IIII, rather than the more usual Roman ``IV. However, despite frequent exposure to such clocks, French and Richards (1993) found that subjects who were asked to draw such a clock from memory typically represented the four on the clock as MEMORY, 1996, 4 (1), 4958 Requests for reprints should be sent to Anne Richards, Department of Psychology, Birkbeck College, University of London, Malet Street, London, WC1E 7HX, UK. Thanks are due to Colin Latham and Sally Beales for assistance with data collection and to Ernest Samuel for producing the images used in Experiment 2. # 1996 Erlbaum (UK) Taylor & Francis D o w n l o a d e d
B y :
[ T e l f e r ,
M a x i n e ]
A t :
0 5 : 3 5
1 3
S e p t e m b e r
2 0 1 0 ``IV rather than the correct ``IIII. Warning subjects that they would be required to draw the clock from memory made no difference to the findings. Subjects who were asked to copy the clock while it remained in full view did not make this mistake. Schema theory was invoked by French and Richards (1993) to account for the findings, as this appeared to offer the most parsimonious explanation. Schemas are cognitive structures representing organised knowledge about concepts, events, situations, etc. Schemas are based on conceptually driven processes (as opposed to data-driven processes) which allow sense to be made from impoverished or ambiguous information. Schemas are used to assist in interpreting the world efficient ly by guiding our attention, and encoding and retrieving information. It is almost certainly the case that schemas are only invoked when needed. The representation of the number four on a clock face is not directly relevant for telling the time. The location of the figure is much more relevant. French and Richards argued that subjects in the two memory conditions examined a couple of the numbers of the clock face in order to note the general style and form of the numbers, but then went on to rely on their schematic knowledge to fill in the details during the drawing of the clock. As the actual numbers on a clock face are less important than their location, when we come to try and remember how a Roman numeral clock face looks, we will invoke our general schema for Roman numerals. The representation of the figures in the Roman numeral counting schema is obviously very important, and the four will be represented as ``IV within this schema. It was also suggested that the subjects in the copy condition used a cognitively economical strategy to draw the clock which did not involve subjects using their schematic knowledge. Subjects in this condition could simply glance at the clock while drawing it. If it is the case that subjects in the copy condition drew the clock without invoki ng their schematic knowledge for Roman numerals, then a subsequent recall task should show that when required to use schematic information, they shoul d recall the four incorrectly, i.e. as ``IV. On the other hand, if those few subjects in the two memory conditions who actually represented the four correctly had consciously noticed that it was represented as ``IIII on the clock face, they should be more likely to recall the number four correctly, i.e. as ``IIII. In Langers (1978, 1989) terms, subjects in the copy condition could copy the clock ``mindlessly, failing to notice the apparent anomaly. The present experiment examined this possibility by requiring subjects to draw the clock in the first phase of the experiment, and then to recall the number on the clock face by writing down how the numbers actually appeared on the clock face. In summary, the present experiment is testing two hypotheses. First, it is predicted, in line with French and Richards (1993) study, that subjects in the copy condition will be more likely to represent the four correctly (as ``IIII) on 5 0 R I C H A R D S , F R E N C H , H A R R I S D o w n l o a d e d
B y :
[ T e l f e r ,
M a x i n e ]
A t :
0 5 : 3 5
1 3
S e p t e m b e r
2 0 1 0 a drawing compared to the two memory groups (forewarned and surprise memory conditions). Second, it is predicted that of those subjects who correctly represent the four as ``IIII on the drawing task, the ones from the copy condition will be more likely to fail a subsequent recall task compared to those from the memory groups. E X P E R I M E N T 1 M e t h o d Subjects. A total of 93 subjects participated in the experiment, but five were excluded because they failed to fulfil the criteria to be outlined shortly. The final group comprised 88 subjects (20 males) who were all first year undergraduates at the University of Hertfordshire studying psychology, either as a single honour s degree or as part of a combined studies programme. The mean age of the group was 20.88 years (SD = 5.15). Apparatus. The clock used in this study was the same as that used by French and Richards (1993). It was a standard clock made by Samuel Bishop of London. The clock had clear Roman numerals on the face with the four represented as ``IIII. The clock was set at ten past seven. Procedure. Subjects were run in groups of approximately 10 or 11 per session in order to ensure that all subjects could see the clock. The procedure followed French and Richards (1993) with subjects being randomly allocated to one of three conditions as follows: (i) Surprise memory condition. Subjects were told, ``I am going to show you a clock which I want you to examine visually for one minute. The clock was then removed from view and subjects were issued with pencils, erasers, and paper (this took exactly two minutes). Subjects were asked to draw the clock as accurately as possible from memory and informed that they would be allowed six minutes to do so. Subjects were warned after five minutes that only one minute remained. (ii) Forewarned memory condition. Subjects were told, ``I am going to show you a clock which I want you to examine visually for one minute. Then I will ask you to draw it from memory. The clock itself will be removed. You will be allowed six minutes. Following the one-minute examination period, the clock was removed from view and subjects were issued with pencils, erasers, and paper (this took exactly two minutes). Subjects were allowed six minutes to draw the clock, and were given a warning after five minutes that only one minute remained. (iii) Copy condition. Subjects were issued with pencils, paper, and erasers in advance, and the clock remained in full view for the six-minute copying period. These subjects were told, ``I would like you to draw a picture of this clock. You C L O C K - W A T C H I N G 5 1 D o w n l o a d e d
B y :
[ T e l f e r ,
M a x i n e ]
A t :
0 5 : 3 5
1 3
S e p t e m b e r
2 0 1 0 have six minutes. Subjects were given six minutes for this task, and warned after five minutes had elapsed. At the end of the drawing period subjects in all groups received the following instructions: Please now write on the rear of your sheet {subject group and day}, and your age and gender. Once you have done this, will you please write down how the numbers for the hours were represented on the clock face? That is, will you please write down the numbers themselvesi.e. from 1 to 12in the exact form that they were written on the clock face? {In each case, the last two sentences were repeated for emphasis. Further, stress was placed on the emphasised words.} The experimenter ensured that no subject turned over the paper to check on how they had drawn the clock earlier. R e s u l t s Subjects were excluded from the analysis if (a) the clock was drawn with the four not represented as ``IIII or ``IV, or (b) subjects did not recall the four as either ``IIII or ``IV. The data from the 88 subjects who produced useable data are presented in Table 1 (the same results were obtained if the full data set was analysed). There was a significant association between condition and the way in which the four was represented in the drawings (w 2 = 47.05, df = 2, P <0.001), thus strongly supporting the first hypothesis. All of the subjects in the surprise memory condition and the forewarned memory condition who correctly represented the four as ``IIII in their drawings also got it correct on the rear of their sheets. However, of the 28 subjects in the copy condition who correctly represented the four as ``IIII in their drawings, 16 got it wrong on the back of their sheets (i.e. put ``IV rather than ``IIII). This association was significant (Fishers Exact Test, P <0.01; see Table 2), supporting the second hypothesis. T A B L E 1 E x p e r i m e n t 1 : D r a w i n g T a s k SMT FMT CT IV 25 22 2 IIII 2 9 28 Subjects representing four as ``IV and ``IIII under the three conditions of Experiment 1: surprise memory task (SMT), forewarned memory task (FMT), and copy task (CT). 5 2 R I C H A R D S , F R E N C H , H A R R I S D o w n l o a d e d
B y :
[ T e l f e r ,
M a x i n e ]
A t :
0 5 : 3 5
1 3
S e p t e m b e r
2 0 1 0 D i s c u s s i o n The present experiment has clearly replicated that of French and Richards (1993). Indeed, the results here are even stronger, with a much lower proportion of those in the two memory conditions representing the four as ``IIII (around a third of those in the memory groups in French and Richards study had done so). In addition to this, it appears that some of the subjects in the copy condition were performing the task ``mindlessly as over half of these subjects (16 out of 28) recalled the number four as ``IV despite having correctly drawn ``IIII in the earlier phase of the experiment. All of the subjects in the memory conditions who drew the number four correctly in the first phase of the experiment subsequently recalled the number correctly. A parsimonious explanation for the performance of those subjects in the copy condition who were right-then-wrong (drew the four correctly but recalled it incorrectly) is in terms of schema theory. These subjects would have copied the clock ``mindlessly, without invoking any schema. However, when they came to try and recall the nature of the numbers on the clock, they would have had to rely on their schematic knowledge of Roman numerals, and consequently misrepresented the number four. Subjects who drew the clock correctly and recalled it correctly may have noticed the anomalous number four during the drawing of the clock, and then gone on to recall the four correctly. An alternative explanation for the performance of the right-then-wrong subjects in the copy condition is that they did notice that the number four was represented as ``IIII, copied it down correctly, but then forgot it by the time the recall task was performed. It was therefore decided to run a second experiment using a more sensitive measure of memory, that of recognition. The recognition paradigm is a more implicit test of memory than is the recall paradigm, and might be a more sensitive measure of what memory representations persist for the surface forms of the numerals. It is possible that some of the subjects who drew and recalled the clock correctly were already aware that clocks with Roman numerals on their face typically have the four represented as ``IIII. For these subjects, there would T A B L E 2 E x p e r i m e n t 1 : S u b s e q u e n t R e c a l l T a s k SMT/FMT CT IV 0 16 IIII 11 12 Subsequent recall by subjects who correctly represented the four as ``IIII under the conditions of Experiment 1: surprise memory task and forewarned memory task combined (SMT/ FMT) and copy task (CT). C L O C K - W A T C H I N G 5 3 D o w n l o a d e d
B y :
[ T e l f e r ,
M a x i n e ]
A t :
0 5 : 3 5
1 3
S e p t e m b e r
2 0 1 0 have been no incongrui ty with their schemas. This is unlikely to provide a complete explanation for the results found, as subjects were randomly allocated to each of the three groups, and it is unlikely that such subjects were disproportionately allocated to the two memory conditions. However, the second study took this into account by asking subjects to state how the number four is usually represented on a Roman numeral clock face. For Experiment 2, subjects were again allocated to the three drawing conditions but were then presented with a recognition test. The recognition task involved subjects having to decide which of two computer-generated pictures of the clock (one with the four correctly represented as ``IIII and one with the four modified to ``IV, see Fig. 1) had been presented. As this was a forced-choice paradigm, a confidence rating scale was incorporated so that subjects who were guessing could be eliminated. In addition to this, at the end of the experiment, subjects were asked to write down how the number four was usually represented on clocks that have Roman numerals on their face. E X P E R I M E N T 2 M e t h o d Subjects. A total of 418 subjects participated in the study, but 171 were excluded for the reasons to be outlined shortly. The final group comprised 247 subjects (58 males) who were all first year students from the Psychology Departments of Birkbeck College and Goldsmiths College. The mean age of the group was 28.11 years (SD = 9.31). Procedure. The experiment consisted of three stages. The first stage involved subjects being allocated to one of three conditions (copy, forewarned memory, and surprise memory) in the same way as for Experiment 1. Subjects were required to reproduce a drawing of the clock. All drawings were collected before the next phase commenced. In Phase II subjects were presented with two pictures of the clock. The pictures were presented face down and subjects were only allowed to look at them two minutes after the end of the drawing task. These two pictures were identical apart from the representation of the number four (see Fig. 1). On one clock the four was represented correctly as ``IIII, and on the other the four was represented incorrectly as ``IV. Half of the subjects received pairs of pictures in which the correct choice was on the left, and half received pairs of pictures in which the correct choice was on the right. Subjects were instructed as follows: Please look carefully at the two pictures on the sheet, one is marked A and the other is marked B. Please decide which one of these clocks is more similar to the clock you saw earlier, and write your answer (A or B) in the space provided. 5 4 R I C H A R D S , F R E N C H , H A R R I S D o w n l o a d e d
B y :
[ T e l f e r ,
M a x i n e ]
A t :
0 5 : 3 5
1 3
S e p t e m b e r
2 0 1 0 F I G . 1 . Pictures of clocks used in the recognition phase of Experiment 2. One clock has the four represented as ``IIII and the other has the four represented as ``IV. 5 5 D o w n l o a d e d
B y :
[ T e l f e r ,
M a x i n e ]
A t :
0 5 : 3 5
1 3
S e p t e m b e r
2 0 1 0 Please then indicate how confident you are about your decision. If you are extremely confident, then circle the 7, if you are unsure about your response, then circle the 1. The rating scale presented at the bottom of the response sheet ranged from 1 (not at all confident) to 7 (extremely confident). All response sheets were collected in before the final phase began. Finally, subjects were asked, ``How is the number four usually represented on a clock which has Roman numerals on its face?. R e s u l t s Subjects were excluded from the analyses if they omitted any of the three stages outlined. In addition to this, subjects were excluded if they failed to draw either ``IIII or ``IV during Stage 1 (some subjects drew ``VI and were thus excluded). Subjects were excluded if they had a confidence rating below 4 during the recognition stage. This was to exclude subjects who were guessing as to which clock had been presented earlier. Finally, subjects whose performance indicated that they were aware that clocks with Roman numerals usually had the four represented as ``IIII were also excluded. The final data set consisted of 247 subjects. The analyses of the full data set (418 subjects) produced the same effects as for the subgroup. Table 3 contains the numbers of subjects representing the four as ``IV and ``IIII for each of the three conditions. There was a significant association between condition and the representation of the number four on the clock face (w 2 = 112.94, df = 2, P <0.001). Further analysis showed that there was no association between the representation of the four and the two memory conditions (w 2 = 0.38, df = 1, ns). However, when the two memory conditions were jointly compared with the copy condition, there was a significant association with the representation of the four (w 2 = 112.66, df = 1, P <0.001). These findings are entirely consistent with those of Experiment 1, and French and Richards (1993). T A B L E 3 E x p e r i m e n t 2 : D r a w i n g T a s k SMT FMT CT IV 55 59 6 IIII 20 17 90 Subjects representing four as ``IV and ``IIII under the three conditions of Experiment 2: surprise memory task (SMT), forewarned memory task (FMT), and copy task (CT). 5 6 R I C H A R D S , F R E N C H , H A R R I S D o w n l o a d e d
B y :
[ T e l f e r ,
M a x i n e ]
A t :
0 5 : 3 5
1 3
S e p t e m b e r
2 0 1 0 An analysis was performed on the recognition data of those subjects who correctly represented the number four in Stage 1 (see Table 4). This analysis revealed that there was a significant association between condition and whether the correct or incorrect clock was recognised (w 2 = 6.17, df = 2, P <0.05). Again, further analyses showed that there was no association between the two memory conditions and the recognition of the clock (Fishers exact test, ns), but if the two memory conditions are collapsed and compared with the copy condition there was a significant association (w 2 = 4.90, df = 1, P <0.05). D i s c u s s i o n Experiment 2 has replicated the basic effect described by French and Richards (1993) and Experiment 1 here, that when subjects drew the clock from memory, the majority mistakenly drew ``IV rather than the correct ``IIII, whereas subjects who merely copied the clock tended not to make this mistake. Analysis of the subsequent recognition performance of those subjects who drew the number four correctly in the first phase of the experiment showed that there was a significant association with condition. Of those subjects in the copy condition drawing the clock correctly, 26% subsequently recognised it incorrectly, as compared to 8% in the memory conditions. This finding supports the idea that several subjects in the copy condi tion drew the clock ``mindlessly. In addition to this, the possibility that subjects in the memory conditions were drawing and recognising the number four correctly because they already knew how fours are represented on clocks with Roman numerals was ruled out, as such subjects in the present experiment were excluded. It appears that those subjects who recognised the clock correctly must have noticed the incongruous number four during the first stage of the experiment. Whether subjects would update their schema for Roman numerals on clock faces as a result of this experience is unknown, although it is likely that such subjects would check future clocks for their representation of the four as a result of their experience. T A B L E 4 E x p e r i m e n t 2 : S u b s e q u e n t R e c o g n i t i o n T a s k SMT FMT CT IV 3 0 23 IIII 17 17 67 Subsequent recognition by subjects who correctly represented the four as ``IIII under the three conditions of Experiment 2: surprise memory task (SMT) forewarned memory task (FMT) and copy task (CT). C L O C K - W A T C H I N G 5 7 D o w n l o a d e d
B y :
[ T e l f e r ,
M a x i n e ]
A t :
0 5 : 3 5
1 3
S e p t e m b e r
2 0 1 0 G E N E R A L D I S C U S S I O N The two experiment s both replicate the finding reported by French and Richards (1993), showing that subjects drawing the clock from memory are significantly more likely to draw the four incorrectly than are subjects copying the clock. The effect of drawing the clock on a later memory task was related to whether subjects drew the clock from memory or simply copied it. There was a greater proportion of subjects in the copy condition as compared to the memory conditions who appeared to have drawn the clock ``mindlessly (i.e. they drew it correctly, but recalled or recognised it incorrectly). For those subjects who produced a correct response on the memory task, this appeared to be due to them noticing the incongruity between their schematic knowledge and the clock, rather than them already knowing how fours are represented on clocks with Roman numerals. Of those subjects correctly drawing the number four in the copy condition, 57% of subjects in Experiment 1 incorrectly recalled the number four, whereas in Experiment 2 only 26% incorrectly recognised the number four. This could be due to the increased sensitivity of the recognition paradigm used in Experiment 2. This paradigm is a more ``implicit test of memory, and it could be the case that subjects had noticed the incongruity in Experiment 1 but then failed to recall it, whereas this was not the case for Experiment 2. An alternative explanation is that the recognition paradigm allows a higher degree of guessing by subjects, with differences in percentages between Experiments 1 and 2 being due to the extra variability in Experiment 2. This problem was countered to a degree by excluding subjects who were not confident about their recognition decision. However, both experiments have supported the hypotheses outlined in the introduction, and support is thus offered for an explanation of the results in terms of schema theory. Manuscript received 28 February 1993 Manuscript accepted 9 September 1994 R E F E R E N C E S French, C.C., & Richards, A. (1993). Clock this! An everyday example of a schema-driven error in memory. British Journal of Psychology, 84, 249253. Langer, E.J. (1978). Re-thinking the role of thought in social interaction. In J.H. Harvey, W.I. Ickes, & R.F. Kidd (Eds.), New directions in attribution research (Vol. 2, pp. 3558). Hillsdale, NJ: Lawrence Erlbaum Associates Inc. Langer, E.J. (1989). Minding matters. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 22, pp. 137173). New York: Academic Press. 5 8 R I C H A R D S , F R E N C H , H A R R I S D o w n l o a d e d