Misleading postevent information: Testing parameterized models of integration in memory

A classic series of experiments by Loftus, Miller and Burns (1978) showed that a person's recollection of an event can be changed by misleading postevent information. Several hypotheses accounting for this effect have been proposed. Loftus' hypothesis of destructive updating claims that the original memory is destroyed by the postevent information. The coexistence hypothesis asserts that the older memory survives but is rendered inaccessible through a mechanism of inhibition or suppression. The non-conflict hypothesis simply accounts for the effect by claiming that subjects can only be misled if they did not encode or if they forgot the original event. These three hypotheses were modelled with the help of all-or-none probabilistic event trees. An experiment was conducted in order to test the three models and to assess parameter values. The experiment followed the classic Loftus paradigm. We suggested to some subjects that they had seen a stopsign, whereas in fact they had seen a traffic light. The misleading postevent information resulted in poorer reproduction of traffic light. Later, all subjects were asked whether they could remember the color of the traffic light, even if they believed they had seen a stopsign. The results showed that subjects who received the misleading post-event information were at least as good at recalling the color of the traffic light as subjects who did not receive misleading information. The no-conflict model accounts well for the obtained results, although the two other, less parsimonious, models cannot be entirely rejected.     

Composites, compromises, and CHARM: what is the evidence for blend memory representations?

Metcalfe's (1990) distributed memory model simulates many misinformation effects by assuming representations that superimpose information from multiple sources. In the present article, two types of evidence are reviewed for such "blend" representations: composite recollections, including items from both the original and postevent sources (e.g., a previously seen intersection is remembered with a subsequently suggested stop sign), and compromise recollections, including features that cannot be exclusively associated with either source (e.g., a green car that was later suggested to be blue is remembered as bluish green). The considerable evidence for composite recollections provides little support for blend representations. Compromise recollections, though seemingly more persuasive, are both rare and interpretable without postulating blend representations. Speculation is made about potential findings that would support blend representations.     

The fate of memory: comment on McCloskey and Zaragoza

McCloskey and Zaragoza (1985) argue that misleading postevent suggestions do not affect the availability of originally encoded information. Their hypothesis stems from empirical work using a modified paradigm in which no effect of postevent information is observed. Although their "no impairment" hypothesis is plausible, careful consideration of the predictions of their experimental test suggests that it may be insufficiently sensitive to reveal the impact of postevent information. A small effect of postevent information can be observed when their paradigm is repeated with a more sensitive recognition test. McCloskey and Zaragoza's no impairment hypothesis is also difficult to reconcile with numerous reports of "blend" memories that reflect a compromise between the original and postevent information.     

Memory impairment and source misattribution in postevent misinformation experiments with short retention intervals

The four experiments reported here provide evidence that (1) misleading postevent suggestions can impair memory for details in a witnessed event and (2) subjects sometimes remember sug-gested details as things seen in the event itself. All four experiments used recall tests in which subjects were warned of the possibility that the postevent information included misleading sug-gestions and were instructed to report both what they witnessed in the event and what was men-tioned in the postevent narrative. Recall of event details was poorer on misled items than on control items, and subjects sometimes misidentified the sources of their recollections. Our re-sults suggest that these findings are not due to guessing or response biases, but rather reflect genuine memory impairment and source monitoring confusions.     

Editing misleading information from memory: Evidence for the coexistence of original and postevent information

When misleading postevent information biases one’s memory for an event, what is the fate of the original, accurate information? One possibility is that the new information and the original information coexist in memory, but that the former is simply more accessible. A second hypothesis suggests that the new information replaces the old, and memory is irreversibly altered. Using various retrieval techniques, Loftus and her associates (Greene, Flynn, & Loftus, 1982; Loftus, 1979a, 1979b) have failed repeatedly in attempts to recover original memories after postevent biasing, a finding that supports the alteration hypothesis. In the present study, postevent biasing was demonstrated in two experiments. In each experiment, some subjects were given a warning that the postevent information had included a few inaccurate details. These subjects were able to edit out the inaccurate details and to recover the original facts when the warning came as much as 45 rain after they had read the misleading information, a result that argues for the coexistence of memories. Successful recovery of the original memories was apparently due to the clarity of the warning and to an improved technique for assessing the retrieval of original memories.     

Dissociating automatic and intentional processes in children's eyewitness memory

Two experiments investigated the contribution of automatic and intentional memory processes to 5- and 8-year-old children's acceptance of misinformation. Children were presented with a picture story followed by misleading postevent details that either were read to participants or were self-generated in response to semantic and perceptual hints. Children were then given a recognition test under 2 instructional conditions. In the inclusion condition children reported whether they remembered items from either of the previous phases. In the exclusion condition children were instructed to exclude postevent suggestions. Children were more likely to accept misled-generate items compared to misled-read items in the inclusion condition, but the opposite was the case under exclusion instructions. Both automaticity and recollection (cf. L. L. Jacoby, 1991) influenced misinformation acceptance, but the role of automatic processes declined with age.     

Color blend retrievals: Compromise memories or deliberate compromise responses?

A total of 978 subjects, in two experiments and two pilot studies, were asked to recognize the color of objects to which they had been exposed in a slide presentation. Taken together, the studies demonstrate that an object’s typical coloration influences recognitions, resulting in retrievals that compromise or blend the actual color and the typical color. In addition, Experiment 2 suggests that color recognitions may be simultaneously influenced by typical knowledge, event information, and postevent information. The findings question the adequacy of both the deliberate compromise and the compromise memory hypotheses as accounts for color blend retrievals. The deliberate compromise hypothesis fails to account for evidence that blend retrievals result because an intact representation of an event is not available at the time of test. The compromise memory hypothesis fails to account for evidence that deliberative processes do occur, and is in need of further exposition given that blend retrievals may be the result of different processes when resulting from typical knowledge and from postevent information. In conclusion, neither hypothesis, in isolation, is able to account for the various possible processes that appear responsible for blend retrievals.     

Role of environmental context in eyewitness memory

We investigated whether varying the environmental context will affect the magnitude of retroactive interference produced by misleading postevent information in an eyewitness memory paradigm. Previous eyewitness memory studies have typically presented the original and misleading information in the same environmental context. In this experiment, the physical contexts in which the original information and the misleading information were presented were varied, a procedure that is more analogous to what usually occurs in real world situations. We tested 288 subjects, half using the original and misleading information in the same encoding context and half using a different context for presenting the two types of information. Memory for the original event was assessed using either the standard recognition test procedure or the modified test developed by McCloskey and Zaragoza (1985). Measures of both recognition accuracy and response latency showed no difference in performance attributable to varying the environmental context. The present data replicate the findings of previous single-context experiments that showed the two recognition test procedures to produce different patterns of results. Thus, environmental context seems to play little role in determining the magnitude of the misleading postevent information effect.     

Memory impairment in the weapon focus effect

Two experiments are reported in which postevent source of misinformation was manipulated within weaponpresent and weapon-absent scenarios. Participants viewed slides depicting either a weapon or a newspaper event and then received either incomplete questioning or a narrative. Both postevent sources contained misleading information about a central and peripheral detail concerning either the weapon or the newspaper scenario. With a modified test in Experiment 1, questioning was found to increase misinformation effects concerning the central item, as compared with a narrative, and more misinformation effects were found for the weapon-peripheral than for the newspaper-peripheral item. In Experiment 2, the participants were more likely to claim to have seen contradictory and additive misinformation about the central item in the slides following questioning, and more contradictory and additive misinformation effects occurred for the weapon-peripheral than for the newspaper-peripheral item. The findings are considered in terms of the effects of both postevent and encoding factors on memory.     

Representations, predictions, and remembrances in CHARM: a reply to Lindsay (1991)

Lindsay (1991), in his comment CHARMed, but not Convinced: Comment on Metcalfe (1990), acknowledged that distributed models of human memory using the construct of a composite memory trace, such as the Composite Holographic Associative Recall Model (CHARM), are able to account for most of the findings within the eyewitness-testimony paradigm. Despite this success, Lindsay found CHARM to be of limited usefulness as a model of eyewitness suggestibility. The reasons stated for this peculiar conclusion are discussed in this rebuttal.     

Misled subjects may know more than their performance implies

Many studies have demonstrated that subjects exposed to misleading postevent information are likely to report the misinformation with confidence on subsequent tests of memory for the event. The purpose of the present studies was to determine whether subjects exposed to misleading postevent information come to believe they remember seeing the misinformation at the original event. A second question addressed by the present studies is whether exposure to misinformation reduces subjects' ability to remember the source of items they witnessed at the original event. In two experiments, subjects viewed a slide sequence depicting an event, were subsequently exposed to misleading information or neutral information about selected aspects of the event, and were later tested on their memory for the source of original and misleading details. The results showed that exposure to misinformation did not lead subjects to believe they remembered seeing the misinformation, nor did it reduce subjects' ability to accurately identify the source of originally seen details. The same pattern of results was obtained whether subjects were tested immediately (Experiment 1) or after a 1-day delay (Experiment 2). Collectively, the results suggest that subjects may report misinformation even if they know they do not remember seeing it.     

The power of the spoken word: sociolinguistic cues influence the misinformation effect

We investigated whether the sociolinguistic information delivered by spoken, accented postevent narratives would influence the misinformation effect. New Zealand subjects listened to misleading postevent information spoken in either a New Zealand (NZ) or North American (NA) accent. Consistent with earlier research, we found that NA accents were seen as more powerful and more socially attractive. We found that accents per se had no influence on the misinformation effect but sociolinguistic factors did: both power and social attractiveness affected subjects' susceptibility to misleading postevent suggestions. When subjects rated the speaker highly on power, social attractiveness did not matter; they were equally misled. However, when subjects rated the speaker low on power, social attractiveness did matter: subjects who rated the speaker high on social attractiveness were more misled than subjects who rated it lower. There were similar effects for confidence. These results have implications for our understanding of social influences on the misinformation effect.     

Recollection in an episodic-like memory task in the rat

Episodic memory in humans is the conscious recollection of a past event. Animal models of episodic-like memory assess the memory for "what"happened, "where" it happened, and either "when" it happened, or in "which" context it happened, although recollection on such tasks is often difficult to measure. Here we present the first evidence of successful recollection of a past event in a rat in a task which is easily performed, requires little training, and is easily adaptable for other commonly used laboratory species.     

Composite holographic associative recall model (CHARM) and blended memories in eyewitness testimony

The idea that compositing or blending occurs in human episodic memory stems from two sources: (a) distributed memory models and (b) studies on the errors that occur in eyewitness testimony. These two traditions of research--theoretical and empirical--have been independent and distinct. Here, data from the eyewitness testimony paradigm are simulated by the distributed model CHARM (Composite Holographic Associative Recall Model). Of focal concern are Loftus's studies, which have been interpreted in favor of the blending hypothesis, and McCloskey and Zaragoza's studies, which have been interpreted as refuting Loftus's position. Both of these seemingly contradictory results, as well as recent findings with yes/no recognition, fall out of the model. Finally, the model predicts empirically found color shifts and provides specifications for when blends and memory impairments will and will not be expected.     

The power of the spoken word: Sociolinguistic cues influence the misinformation effect

We investigated whether the sociolinguistic information delivered by spoken, accented postevent narratives would influence the misinformation effect. New Zealand subjects listened to misleading postevent information spoken in either a New Zealand (NZ) or North American (NA) accent. Consistent with earlier research, we found that NA accents were seen as more powerful and more socially attractive. We found that accents per se had no influence on the misinformation effect but sociolinguistic factors did: both power and social attractiveness affected subjects' susceptibility to misleading postevent suggestions. When subjects rated the speaker highly on power, social attractiveness did not matter; they were equally misled. However, when subjects rated the speaker low on power, social attractiveness did matter: subjects who rated the speaker high on social attractiveness were more misled than subjects who rated it lower. There were similar effects for confidence. These results have implications for our understanding of social influences on the misinformation effect.     

The integration of familiarity and recollection information in short-term recognition: modeling speed-accuracy trade-off functions

Dual process models postulate familiarity and recollection as the basis of the recognition process. We investigated the time-course of integration of the two information sources to one recognition judgment in a working memory task. We tested 24 subjects with a response signal variant of the modified Sternberg recognition task (Oberauer, 2001) to isolate the time course of three different probe types indicating different combinations of familiarity and source information. We compared two mathematical models implementing different ways of integrating familiarity and recollection. Within each model, we tested three assumptions about the nature of the familiarity signal, with familiarity having (a) only positive values, indicating similarity of the probe with the memory list, (b) only negative values, indicating novelty, or (c) both positive and negative values. Both models provided good fits to the data. A model combining the outputs of both processes additively (Integration Model) gave an overall better fit to the data than a model based on a continuous familiarity signal and a probabilistic all-or-none recollection process (Dominance Model).     

Event-related potential (ERP) evidence for varied recollection during source monitoring

The source monitoring framework (SMF; M. K. Johnson, S. Hashtroudi, & D. S. Lindsay, 1993) posits that source monitoring can be supported by varying degrees of recollection. Source judgments were made for words heard at study (male or female voice) followed by remember/know (RK) judgments in order to assess differences in degrees of recollection during source monitoring (J. L. Hicks, R. L. Marsh, & L. Ritschel, 2002). The neural correlate of recollection was examined by contrasting event-related potentials (ERPs) for R and K responses that followed accurate source judgments. The recollection ERP effect was greater for R than K responses but greater for K responses than correct rejections, which reinforces the behavioral evidence that indicates source monitoring can be supported by varying degrees of recollection. A robust late posterior negativity ERP effect was present and began earlier for accurate source judgments given subsequent R than K responses. This is interpreted as reflecting systematic decision processes that are described by the SMF.     

The eyewitness suggestibility effect and memory for source

We examined the possibility that eyewitness suggestibility reflects failures of the processes by which people normally discriminate between memories derived from different sources. To test this hypothesis, misled and control subjects were tested either with a yes/no recognition test or with a "source monitoring" test designed to orient subjects to attend to information about the sources of their memories. The results demonstrate that suggestibility effects obtained with a recognition test can be eliminated by orienting subjects toward thinking about the sources of their memories while taking the test. Our findings indicate that although misled subjects are capable of identifying the source of their memories of misleading suggestions, they nonetheless sometimes misidentify them as memories derived from the original event. The extent to which such errors reflect genuine memory confusions (produced, for example, by lax judgment criteria) or conscious misattributions (perhaps due to demand characteristics) remains to be specified.     

The lights are on but no one’s home: Meta-awareness and the decoupling of attention when the mind wanders

In a recent review, we suggested that an important aspect of mind-wandering is whether participants are aware that they are off task (Smallwood & Schooler, 2006). We tested this hypothesis by examining the informationprocessing correlates of mind wandering with and without awareness in a task requiring participants to encode words and detect targets with either a high or a low probability. Target detection was measured via response inhibition. Mind wandering in the absence of awareness was associated with a failure to supervise task performance, as indicated by short RTs, and was predictive of failures in response inhibition. Under conditions of low target probability, mind wandering was associated with a relative absence of the influence of recollection at retrieval. The results are consistent with the notion that mind wandering involves a state of decoupled attention and emphasizes the importance of meta-awareness of off-task episodes in determining the consequences of these mental states.