Context Reinstatement Effects in Children's Cued Recall of Strongly and Weakly Associated Word Pairs

Context Reinstatement Effects inChildren’s Cued Recall of Strongly and Weakly Associated Word Pairs

Paul M. Dietze, Stefanie J. Sharman, Martine B. Powell and Donald M. ThomsonDeakin University, Melbourne, Australia

Typically, asking people to reinstate the context of events increases their recallof those events; however, research findings have been mixed with children. Wetested whether the principle underlying context reinstatement applies tochildren as it does to adults. This underlying principle, encoding specificity,suggests that the greater the overlap between study context cues and retrievalcontext cues, the more information that people should recall. In the currentexperiment, four age groups (7-year-olds, 9-year-olds, 11-year-olds and adults)took part in an encoding specificity procedure. At study, participants saw cue–target word pairs in which the cue word was either a strong or a weak associateof the target word (e.g., ice–COLD; blow–COLD). During an immediate cuedrecall test, participants were presented with the same strong or weak cuewords and new, extra-list cue words. Overall, children and adults recalled moretargets when they were presented with the same cue words at study and test,regardless of whether the cues were strong or weak. This finding suggests thatencoding specificity applies to children as well as adults. We discuss theimplications of these results.

❚ Keywords: cued recall, children, context reinstatement

In many situations, it is essential for children to provide accurate descriptions ofevents, particularly events to which they have been the only witness. One methodoften used to encourage accurate, detailed recall is to ask them to reinstate the contextin which the original event was experienced (for reviews, see Goodman & Melinder,2007; Miller & Fremouw, 1995; Salmon, 2001). Context reinstatement can take manyforms: people may be taken back to the scene of an event (physical contextreinstatement), they may be asked to imagine the scene of an event (mental contextreinstatement), or they may be provided with cues that help them to recall an event(cued context reinstatement). Context reinstatement increases recall of events becauseit increases the number of available cues for people to use to access their memories ofthe events. Our interest in this area has arisen from the field of investigativeinterviewing of children and the effect of mental context reinstatement.

Mental context reinstatement refers to interviewer instruction that encourageswitnesses to reinstate both the external (environmental) and internal (personal)

Address for correspondence: Dr. Stefanie Sharman, School of Psychology, Deakin University, 221 Burwood Hwy, Burwood VIC 3125, Australia. Email: [email protected]

The Australian Educational and Developmental PsychologistVolume 28 | Issue 2 | 2011 | pp. 91–100 | DOI 10.1375/aedp.28.2.91

❚ 91

contexts that existed when the events were experienced. It is the core technique of theCognitive Interview (Geiselman, 1988; Geiselman, Fisher, MacKinnon, & Holland,1986), which distinguishes this particular interview from other interview techniques.Mental context reinstatement is designed to maximise the feature overlap between theinterview session and the original event, which — according to the principle ofencoding specificity (Tulving & Thomson, 1973) — should facilitate recall. Theencoding specificity principle states that ‘what is stored is determined by what isperceived and how it is encoded, and what is stored determines what retrieval cues areeffective in providing access’ (Tulving & Thomson, 1973, p. 353). Therefore, thegreater the feature overlap between encoding and retrieval environments, the better aperson’s recall performance (Flexser & Tulving, 1978).

Results from research investigating the effects of mental context reinstatement onchildren’s memory have been mixed. In one of our previous studies, mental contextreinstatement improved children’s recall (Dietze & Thomson, 1993). In this study, 6-year-olds, 11-year-olds and adults watched a five minute film. After a 1-hour interval,they were interviewed about the film in one of three conditions: free recall, mentalcontext reinstatement or specific questions about the film. The mental contextreinstatement instruction increased participants’ recall of correct details withoutincreasing their recall of incorrect details, compared to the other two conditions.However, many other studies have shown no improvement in children’s memory(Hershkowitz, Orbach, Lamb, Sternberg, & Horowitz, 2001, 2002; Memon, Cronin,Eaves, & Bull, 1995; Memon, Holley, Wark, Bull, & Köhnken, 1996). In fact, a review ofthe literature revealed that only three out of seven studies demonstrated that mentalcontext reinstatement increased children’s recall (Dietze, Powell, & Thomson, 2010;Dietze & Thomson, 1993; Hayes & Delamothe, 1997).

Given these inconsistent findings from investigations of mental reinstatementinstructions on children’s recall, it is difficult to draw conclusions about its effectiveness.In fact, it is possible that the encoding specificity principle does not apply to children inthe same way that it does to adults. We decided to take a step back and test thishypothesis to determine whether the theory on which the mental reinstatement ofcontext is based — the encoding specificity principle — also applies to children. Toinvestigate this hypothesis, we used Thomson and Tulving’s (1970) originalexperimental procedure, in which physical context reinstatement was used rather thanmental context reinstatement. The encoding specificity principle refers to any contextcues — whether they are physical or mental cues. Therefore, if we can demonstrate thatreinstatement of physical context cues increases children’s recall in a manner similar tothe way in which reinstatement of physical context cues increases adults’ recall, we canassume that the principle of encoding specificity applies to children as it does to adults.Furthermore, given that reinstatement of mental cues also increases adults’ recall, itshould be reasonable to assume reinstatement of mental cues should also increasechildren’s recall.

In Thomson and Tulving’s (1970) experiment, adult participants learned cue–target word pairs at study. The cue words were either strong or weak associates oftarget words; for example, for the target word ‘black’ the strong associate was ‘white’and the weak associate was ‘train’. During a later cued recall test, participants werepresented with either the same strong or weak cue words that had been presented atstudy, or they were presented with new cue words. Participants recalled more targetwords when the same cues were presented during study and test, regardless of whether

Paul M. Dietze, Stefanie J. Sharman, Martine B. Powell and Donald M. Thomson

The Australian Educational and Developmental Psychologist92 ❚

those cues were strong or weak associates. Therefore, memory performance dependedupon reinstatement of the context in which material was learned, which demonstratedthe principle of encoding specificity.

In our experiment, we recruited participants from four age groups: 7-year-olds, 9-year-olds, 11-year-olds and adults. We included a group of adult participants for tworeasons. First, we wanted to ensure that we could replicate Thomson and Tulving’s(1970) findings. Second, we included adults as a comparison group to determinewhether context reinstatement cues were less effective with children. Participants weregiven the same or different weakly and strongly related cues for the target words at studyand test. We made a number of predictions. First, we predicted that participants shouldrecall more target words when the study and test cues matched, regardless of whetherthey were weak or strong associates of the target words. Second, we predicted that —when the study and test cues matched — participants would recall more targets tostrongly associated cue words than weakly associated cue words. Third, we predictedthat there should be an interaction between study and test cues and age, such thatchildren and adults’ recall should vary with the types of cues they were provided.

MethodPARTICIPANTS AND DESIGNEighty participants took part in the experiment; 20 in each of the four age groups. The7, 9 and 11-year-old participants came from a primary school in Melbourne,Australia. The 7-year-olds were recruited from grades 1 and 2 (mean age 6 years 11months), the 9-year-olds from grades 3 and 4 (mean age 8 years 7 months), and the11-year-olds were recruited from grades 5 and 6 (mean age 10 years 8 months). Theadult participants (i.e., over 18 years) were undergraduate students and junior staff.The experiment used a 4 (age: 7 years, 9 years, 11 years, adults) x 2 (study cue: weak,strong) x 2 (test cue: weak, strong) mixed design. Age was the between-participantsfactor; study cue and test cue were the within-participants factors. Ethical approvalwas granted for the study before any testing took place.

MATERIALS AND PROCEDURE

Pilot. Before conducting the experiment, a pilot study determined whether presentingthe words visually or auditorily affected people’s recall. Auditory presentation wasdesigned to remove any differences in reading ability of the different age groups. Inthe pilot study, 20 adults were presented the words visually; 20 were presented thewords auditorily in the procedure described below. There were no differences in recallbetween the visual and auditory group, F(1, 38) < 1, p > .05. Thus, auditorypresentation of words was used for all groups in the experiment. Participants werepresented with two lists during the experiment; the first list was a practice list.

Experiment. Participants were asked to listen carefully to words presented onaudiotape for a later memory test. They were given a list of 24 cue–target word pairs.Each word pair was separated by 2 seconds and cue words were presented immediatelybefore target words. Target words were presented considerably louder than cue words,which should have helped participants to distinguish targets from cues. Participantswere asked to try to remember the ‘loud’ words and also to pay attention to the ‘soft’words as these soft words may help them to remember the loud words. Half of theword pairs were strong associates (e.g., ice–COLD); the other half were weak

Context Reinstatement

The Australian Educational and Developmental Psychologist ❚ 93

associates (e.g., blow–COLD). Lists were counterbalanced so that targets wereassociated with strong and weak cues equally often. For the practice list, the associateswere generated by the experimenter. For the two experimental lists, associates weredrawn from the Palermo and Jenkins (1964) word association norms and the Palermoand Jenkins (1966) oral word association norms for children in grades 1–4.

Immediately after the study phase, participants’ recall was tested. They werepresented with a list of 24 cue words on audiotape at a rate of one word per 5 seconds.For the practice list, these cues were the same cue words that were paired with thetarget words at study. For the experimental lists, half of the cue words were the sameas those presented at study. Half of these cues were strong associates of the targetwords and half were weak associates of the target word. The other half of the cuewords did not appear at study. That is, if a target was paired with a strong associate atstudy, a weak associate was presented at test and vice-versa. This procedure meant thatthere were two versions of the test cue list for each of the two versions of the study list,producing a total of four test cue lists.

Participants were presented with one of the four test lists. They listened to each cueword and verbally responded with the target word that the cue was paired with atstudy if they could remember it. The experimenter noted their recall. At the end of thelist of cue words, participants were asked if they could remember any more of theloud words. The experimenter noted any additional words that participants couldremember in the absence of a cue word. The combination of the study cue and testcue variables resulted in the four different conditions of Thomson and Tulving’s(1970) Experiment 3: strong–strong, strong–weak, weak–weak, and weak–strong.There were six items per condition per study list, giving a total of 24 target words.

SCORINGParticipants’ responses were scored as correct if they were any of the target items. Thatis, recalled targets were scored as correct irrespective of whether or not they werepaired with their associated cue words (although it was noted whether each pairingwas correct). For each participant, we calculated separate correct recall scores for eachof the four study–test cue combinations. Participants’ recall of words in response tocues that were not target words were scored as intrusion errors. Those made after thepresentation of all the cue words were categorised as ‘no cue’ intrusion errors.

ResultsCORRECT RESPONSESFigure 1 shows the mean number of correct responses as a function of age for the fourstudy–test cue combinations. A 4 (age) x 2 (study cue) x 2 (test cue) one-way Analysisof Variance (ANOVA) revealed a significant three-way interaction, F(3, 76) = 15.08,MSE = 1.34, p < .001. There was also a significant two-way interaction between studycue and test cue, F(1, 76) = 339.20, MSE = 1.34, p < .001. There were significant maineffects for age, F(3, 76) = 9.45, MSE = 1.63, p < .001, and test cue, F(1, 76) = 124.05,MSE = 1.12, p < .001.

The significant three-way interaction indicated that the interaction between studycue and test cue was different for the four age groups. Analysis of simple interactionsrevealed that the two-way interaction between study cue and test cue was significantfor all four age groups, F(1, 19) ≥ 26.79, p < .001. However, the pattern of this

Paul M. Dietze, Stefanie J. Sharman, Martine B. Powell and Donald M. Thomson

The Australian Educational and Developmental Psychologist94 ❚

interaction changed as a function of age group. Analysis of simple main effectsrevealed that, for all age groups, when participants received a strong study cue, theircorrect recall was higher when they also received a strong cue at test, compared to aweak cue, F (1, 19) ≥ 57.34, p < .001. Similarly, when participants received a weakstudy cue, adults’ and 11-year-olds’ correct recall was higher when they also received aweak cue at test, compared to a strong cue, F(1, 19) ≥ 6.48, p < .05. Although 7-year-olds’ and 9-year-olds’ correct recall was also higher when they received a weak cue attest, compared to a strong cue, these differences failed to reach statistical significance,F(1, 19) ≤ 1.15, p > .10.

Further analyses of simple main effects indicated that, for all age groups, correctrecall was higher when a strong cue at test was preceded by a strong cue, compared toa weak cue, at study, F(1, 19) ≥ 14.02, p < .01. Correct recall was also higher for all agegroups when a weak cue at test was preceded by a weak cue, compared to a strong cue,at study, F(1, 19) ≥ 15.98, p < .001.

Tukey post hoc tests (α = .05) were used to analyse the main effect of age inrelation to all four study–test cue combinations. The analysis revealed an age-relatedincrease in correct recall when the study and test cues matched (strong–strong andweak–weak conditions). In the strong–strong condition, adults and 11-year-olds bothrecalled more than the 7-year-olds. Adults also recalled more than 9-year-olds. In theweak–weak condition, adults recalled more than the 11-year-olds. The main effect ofage in the remaining two study and test cue combinations (strong–weak and weak–strong conditions) was not significant.

Finally, a planned comparison was conducted between the two conditions in whichthe study and test cues matched (strong–strong and weak–weak conditions), collapsedacross the four age groups. It revealed that participants’ correct recall was higher in

Context Reinstatement

The Australian Educational and Developmental Psychologist ❚ 95

FIGURE 1 Mean correct cued recall in each of the four age groups.

the strong–strong than in the weak–weak condition, F(1, 76) = 54.45, MSE = 1.27, p < .001. There was no interaction between correct recall performance in the strong–strong and weak–weak conditions as a function of age, F(3, 76) < 1, p > .05.

CORRECTED SCORESIt is possible that when participants were given strong cues at test they were able toguess the correct target using semantic memory (Thomson & Tulving, 1970). Thus,participants may have high correct recall performance without actually rememberingthe target words from the study phase. To control for this type of guessing,participants’ recall scores were corrected for the conditions in which they were givenstrong cues at test (strong–strong and weak–strong). Their guessing from semanticmemory was estimated using the number of intrusion errors that they made to weaktest cues; that is, the number of strong associate intrusions participants made in thestrong–weak condition (Thomson & Gannon, 1984). For each participant, theirstrong–strong and weak–strong recall scores were corrected by subtracting thenumber of intrusion errors that they made in the strong–weak condition. Thesecorrected scores are displayed in Figure 2, along with the uncorrected scores for theweak–weak condition.

The omnibus analysis of variance revealed a significant interaction between age ofparticipant and the three cue combination conditions, F(6, 152) = 4.75, MSE =1.64, p < .001. There were also significant main effects associated with age, F(3, 76) = 15.68,MSE = 2.79, p < .001, and cue combination, F(2, 152) = 79.11, MSE = 1.64, p < .001.

Inspection of Figure 2 suggests that the two-way interaction resulted from changesin the effect of age as a function of cue combination. Analysis of simple main effectsrevealed a significant effect of age for all cue combinations, F(3, 76) ≥ 4.67, p < .01.

Paul M. Dietze, Stefanie J. Sharman, Martine B. Powell and Donald M. Thomson

The Australian Educational and Developmental Psychologist96 ❚

FIGURE 2Corrected-for-guessing cued recall in each of the four age groups.

The pattern of this age effect, however, changed as a function of cue combination.When the test cue matched the study cue, the pattern was similar to that in Figure 1:recall increased as a function of age. Post hoc tests indicated that in the strong–strongcondition, adults’ corrected scores were higher than 9-year-olds’ scores, which in turnwere higher than 7-year-olds’ scores. Eleven-year-olds’ corrected scores were alsohigher than 7-year-olds’ scores, but there were no significant differences between the11-year-olds’ scores and the remaining two age groups. In the weak–strong condition,adults, 11-year-olds, and 9-year-olds did not differ significantly from each other,although they all had higher corrected scores than the 7-year-olds. As participants’recall was not corrected for guessing in the weak–weak condition, the results were thesame as reported above.

Further analysis of the corrected scores indicated the nature of main effects ofcue combination. Analysis of simple main effects revealed a significant effect of cuecombination for all ages, F(2, 38) ≥ 9.18, p < .001. The pattern of this effect for allage groups was similar. Here, the strong–strong and weak–weak conditions werenot significantly different, but both had higher corrected scores than the weak–strong condition.

INTRUSION ERRORSOverall, the number of intrusion errors was low. Table 1 shows the mean number oferrors for each of the four of the study and test cue combinations. It also shows theerrors participants made in the no cue category, which were errors made after all thetest cues had been presented. The low number of errors in the no cue categoryindicated that few participants made errors, unless in response to a cue word. Thiscategory was not included in any formal analyses.

The omnibus analysis of variance revealed a significant main effect of age, F(3, 76)= 4.60, MSE = 2.19, p < .01. Post hoc tests showed that the 7-year-olds made moreintrusion errors than the other age groups, which did not differ significantly from oneanother. There was also a significant main effect of test cue, F(1, 76) = 30.56, MSE =0.50, p < .001. Participants made more intrusion errors for weak than strong test cues.There were no other significant main effects or interactions.

DiscussionThis experiment examined whether the encoding specificity principle applies tochildren as it does to adults. The answer to this question is yes, however, it depended

Context Reinstatement

The Australian Educational and Developmental Psychologist ❚ 97

TABLE 1Mean Number of Intrusion Errors in the Different Age Groups

Age group

Condition 7 years 9 years 11 years Adults

Strong–strong 0.60 (0.82) 0.25 (0.44) 0.10 (0.31) 0.20 (0.52)

Strong–weak 1.50 (1.73) 0.85 (1.23) 0.50 (1.00) 0.65 (0.75)

Weak–strong 1.05 (1.32) 0.45 (1.12) 0.15 (0.37) 0.50 (0.69)

Weak–weak 1.35 (1.49) 1.00 (0.31) 0.50 (0.83) 0.45 (0.69)

No cue 0.10 (0.31) 0.00 (0.00) 0.00 (0.00) 0.40 (0.77)

Note: Standard deviations are given in parentheses. The ‘No cue’ condition refers to the intrusion errors that participants madeafter they had responded to all the test cues.

on age to some extent. Context reinstatement cues enhanced children’s recall, which isconsistent with previous research demonstrating that extra-list semantic cuesincreased their recall (e.g., Kobasigawa, 1974; Perlmutter, Sophian, Mitchell, &Cavanaugh, 1981). The general pattern of cued recall performance for all age groupswas similar: cued recall with context cues enhanced participants’ memories comparedto cued recall with extra-list semantic cues. This pattern was evident for bothuncorrected and guessing-corrected cued recall scores. These results suggest thatcontext cues are more effective than extra-list semantic cues in cued recall tasks forschool-aged children (just as they are for adults).

While not always significant, the pattern of results, in general, conformed to thepattern expected on the basis of the encoding specificity principle (Tulving &Thomson, 1973). Participants recalled more target words when given the same cues atstudy and test than when they were given different cues. This pattern occurredregardless of whether the cues were strong or weak associates of the target words.Therefore, it would appear that an encoding specificity account of retrieval processescan be generalised across the ages. Our findings extend those reported by Ackerman(1981), who investigated children’s recall of word and picture pairs. In Ackerman’sstudy, the relation between cue and target was emphasised at both study and retrievalby the use of specific questions. These questions highlighted the conceptual oracoustic relationship between objects in the pair (e.g., participants were asked ‘Areboth of these objects fruits?’ for the pair ‘apple–pair’). He also found that participants’recall was increased when they were presented the same objects at study and retrieval.Critically, the findings of the current experiment have demonstrated that, even whenthe relationship between the two items in a pair was not emphasised — such asthrough the use of specific questions, children’s cued recall was increased when theywere given the same contextual cues at encoding and retrieval. Indeed, our results havereplicated Thomson and Tulving’s (1970) original findings with adults and extendedthem to children. It is possible that children’s pattern of recall may have been slightlydifferent had visual rather than auditory word lists been used; however, theexperiment would have been limited by children’s reading ability.

One other important finding from the current experiment was that the contextualcues increased children’s recall without increasing their number of intrusion errors.Overall, the rate of errors (irrespective of age) was very low, which indicates that thecues did not suggest the information that children were required to produce. In otherwords, the contextual cues did not direct children towards other specific, butincorrect, answers. It is important that any technique that increases children’s recall ofaccurate information does not also increase their recall of inaccurate information.Our experiment was not designed to examine context cues and children’ssuggestibility; however, future research might explore the impact of suggestive cues onchildren’s recall.

The results from the current experiment have at least two implications. First, theycontribute to the debate in the literature about whether children are able to mentallyreinstate the context in which they experienced an event to remember more about it.Some studies have found that asking children to mentally reinstate the contextimproved the number of details that they recalled (Dietze et al., 2010; Dietze &Thomson, 1993; Hayes & Delamothe, 1997); others have not (e.g., Herskowitz et al.,2001). The current results showed that context reinstatement cues improved children’srecall, which leads us to speculate that mental context reinstatement cues should also

Paul M. Dietze, Stefanie J. Sharman, Martine B. Powell and Donald M. Thomson

The Australian Educational and Developmental Psychologist98 ❚

improve their recall. It is possible that previous experiments that did not findincreased recall when children were given mental context instructions did not do sobecause of other factors. For example, it is possible that the interviewer’s behavioursand comments affected children’s recall (see, for example, Garven, Wood, Malpass, &Shaw, 1998). It is unlikely that these factors had an impact on recall in the currentexperiment as they were largely social factors that occurred during the interactionbetween the interviewer and the child. In the current experiment, there was verylimited interaction between the experimenter and the participants. Given our findingthat context cues improved children’s recall, mental reinstatement cues — without anyadditional complications caused by social interactions—should also improve recall.

The second implication of the current findings is that they highlight the importantrole of verbal cues during investigative interviews with children in determining theaccuracy and amount of details provided. Finding cues that are nonleading has beenproblematic (see Salmon, 2001, for a review); therefore, any techniques that increaserecall without increasing errors should be encouraged. Interviewers may be able to useverbal cues to increase children’s recall about some details of an event; rememberingthose details may themselves serve as cues to further increase recall.

ReferencesAckerman, B. P. (1981). Encoding specificity in the recall of pictures and words in children and adults.

Journal of Experimental Child Psychology, 31, 193–211.

Dietze, P. M., Powell, M. B., & Thomson, D. M. (2010). Mental reinstatement of context with child witnesses:Does it matter whether context is reinstated ‘out loud’? Psychology, Crime & Law, 16, 439–448.

Dietze, P. M., & Thomson, D. M. (1993). Mental re-instatement of context: A technique for interviewingchild witnesses. Applied Cognitive Psychology, 7, 97–108.

Flexser, A. J., & Tulving, E. (1978). Retrieval independence in recognition and recall. PsychologicalReview, 85, 153–171.

Garven, S., Wood, J. M., Malpass, R. S., & Shaw, J. S. (1998). More than suggestion: The effect ofinterviewing techniques from the McMartin Preschool case. Journal of Applied Psychology, 83, 347–359.

Geiselman, R. E. (1988). Improving eyewitness memory through mental reinstatement of context. In G. M.Davies & D. M. Thomson (Eds.), Memory in context: Context in memory (pp. 245–266). Oxford: Wiley.

Geiselman, R. E., Fisher, R. P., MacKinnon, D. P., & Holland, H. L. (1986). Enhancement of eyewitnessmemory with the Cognitive Interview. The American Journal of Psychology, 99, 385–401.

Goodman, G. S., & Melinder, A. (2007). Child witness research and forensic interviews of youngchildren: A review. Legal and Criminological Psychology, 12, 1–19.

Hayes, B. K., & Delamothe, K. (1997). Cognitive interviewing procedures and suggestibility in children’srecall. Journal of Applied Psychology, 82, 562–577.

Hershkowitz, I., Orbach, Y., Lamb, M. E., Sternberg, K. J., & Horowitz, D. (2001). The effects of mentalcontext reinstatement on children’s accounts of sexual abuse. Applied Cognitive Psychology, 15, 235–248.

Hershkowitz, I., Orbach, Y., Lamb, M. E., Sternberg, K. J., & Horowitz, D. (2002). A comparison ofmental and physical context reinstatement in forensic interviews with alleged victims of sexual abuse.Applied Cognitive Psychology, 16, 429–441.

Kobasigawa, A. (1974). Utilization of retrieval cues by children in recall. Child Development, 45, 127–134.

Memon, A., Cronin, Ó., Eaves, R., & Bull, R. (1995). An empirical test of the mnemonic components ofthe Cognitive Interview. In G. Davies, S. Lloyd-Bostock, M. McMurran, & C. Wilson (Eds.),Psychology, law, and criminal justice: International developments in research and practice (pp. 135–145). England: De Gruyter.

Memon, A., Holley, A., Wark, L., Bull, R., & Köhnken, G. (1996). Reducing suggestibility in child witnessinterviews. Applied Cognitive Psychology, 10, 503–518.

Miller, C. M., & Fremouw, W. J. (1995). Improving the accuracy of adult eyewitness testimony:Implications for children. Clinical Psychology Review, 15, 631–645.

Context Reinstatement

The Australian Educational and Developmental Psychologist ❚ 99

Palermo, D. S., & Jenkins, J. J. (1964). Word association norms: Grade school through college. Oxford:University of Minnesota Press.

Palermo, D. S., & Jenkins, J. J. (1966). Oral word association norms for children in grades one throughfour. Research Bulletin, 60, Pennsylvania: Department of Psychology, Pennsylvania State University,.

Perlmutter, M., Sophian, C., Mitchell, D. B., & Cavanaugh, J. C. (1981). Semantic and contextual cuing ofpreschool children’s recall. Child Development, 52, 873–881.

Salmon, K. (2001). Remembering and reporting by children: The influence of cues and props. ClinicalPsychology Review, 21, 267–300.

Thomson, D. M., & Gannon, R. (1984). The encoding specificity paradigm: A diagnostic memory test. In M. Field, J. Hendy, C. Roberts, & P. Hopkins (Eds.), Brain impairment: Proceedings of the 1984Brain Impairment Conference (pp. 109–118). Melbourne: The Australian Society for the Study ofBrain Impairment.

Thomson, D. M., & Tulving, E. (1970). Associative encoding and retrieval: Weak and strong cues. Journalof Experimental Psychology, 86, 255–262.

Tulving, E., & Thomson, D. M. (1973). Encoding specificity and retrieval processes in episodic memory.Psychological Review, 80, 352–373.

Paul M. Dietze, Stefanie J. Sharman, Martine B. Powell and Donald M. Thomson

The Australian Educational and Developmental Psychologist100 ❚