The role of unconscious memory errors in judgments of confidence for sentence recognition

The present experiment tested the hypothesis that unconscious reconstructive memory processing can lead to the breakdown of the relationship between memory confidence and memory accuracy. Participants heard deceptive schema-inference sentences and nondeceptive sentences and were tested with either simple or forced-choice recognition. The nondeceptive items showed a positive relation between confidence and accuracy in both simple and forced-choice recognition. However, the deceptive items showed a strong negative confidence/accuracy relationship in simple recognition and a low positive relationship in forced choice. The mean levels of confidence for erroneous responses for deceptive items were inappropriately high in simple recognition but lower in forced choice. These results suggest that unconscious reconstructive memory processes involved in memory for the deceptive schema-inference items led to inaccurate confidence judgments and that, when participants were made aware of the deceptive nature of the schema-inference items through the use of a forced-choice procedure, they adjusted their confidence accordingly.     

Long-term memory in amnesia: cued recall, recognition memory, and confidence ratings

We explored what kind of information is acquired when amnesic patients are able to exhibit significant retention on tests of cued recall and recognition memory. Amnesic patients and control subjects attempted to learn sets of sentences. Memory for the last word in each sentence was tested after 1 hr in the case of the amnesic patients, or after 1 to 2 weeks in the case of (delayed) control subjects. Amnesic patients and (delayed) control subjects performed at similar levels on tests of cued recall and recognition memory. Amnesic patients were just as confident of their correct answers as were control subjects. However, amnesic patients were no more disadvantaged than control subjects when they were cued indirectly by presenting paraphrases of the original sentences. These findings demonstrate that the residual knowledge retained by amnesic patients can be as flexible, as accessible to indirect cues, and as available to awareness as the knowledge retained by (delayed) control subjects.     

Accurate identification but no priming and chance recognition memory for pictures in RSVP sequences

In 1969, Potter and Levy reported that recognition memory of accurately perceived RSVP pictures was extremely low, an effect that they attributed to disruption of memory consolidation. Here we report that the repetition of an RSVP picture (72-126 msec/picture) up to 31 times prior to when it became a target had no effect on identification accuracy. At these rates, forced-choice recognition memory was at chance. Single presentations of the pictures outside of the RSVP sequences readily resulted in substantial priming of their identification within the sequences. We offer a neural interpretation of Potter and Levy's explanation, as well as contemporary two-stage accounts of RSVP memory and attentional phenomena, based on the recent finding (Tovee & Rolls, 1995) that most of the information in inferior temporal cells is conveyed in the first 50msec of firing but the cells continue their activity for an additional 350msec. The additional activity, by our account, is required for memory and it is this activity that may be disrupted by attention to the next image during RSVP presentations. The critical factor for priming, if not memory in general, may be attention to the stimulus for a few hundred milliseconds beyond that required for its identification. Single-trial presentations thus manifest robust memory and priming effects - even when the stimulus cannot be identified - whereas RSVP conditions in which the stimulus can be identified result in poor memory.     

Consciousness, control, and confidence: the 3 Cs of recognition memory

The contributions of recollection and familiarity to recognition memory performance were examined using the process dissociation, remember-know, and receiver operating characteristic (ROC) procedures. Under standard test conditions the 3 measurement procedures led to process estimates that were almost identical and to similar conclusions regarding the effects of different encoding manipulations. Dividing attention led to a large decrease in recollection and a smaller, sometimes nonsignificant, decrease in familiarity. Semantic compared with perceptual processing led to a large increase in recollection and a moderate increase in familiarity. Moreover, the results showed that familiarity was well described by classical signal-detection theory but that recollection reflected a threshold process. The convergence observed across the 3 measurement procedures shows that the 3 procedures tap similar underlying processes and that recollection and familiarity differ in terms of conscious awareness, intentional control, and the manner in which they contribute to the shape of response confidence ROCs.     

Visual recognition memory for binary pictures: another look

Recognition memory for matrices of 0s and 1s was examined as a function of the number of elements (complexity) and density of those elements within the matrix. It was found that with greater density and lesser complexity, recognition performance improved. This result contradicts an earlier finding of Green and Purohit, who concluded that the more complicated matrices led to better performance. The present study suggests that their conclusion was erroneous and stemmed from failure to control the physical parameters of density and number of elements in the matrix. Implications for the general problem of recognition memory and eye movements are discussed.     

ROC curves and confidence judgements in recognition memory

Most models of recognition memory rely on a strength/familiarity-based signal detection account that assumes that the processes giving rise to a confidence judgment are the same as those giving rise to an old-new decision. Confidence is assumed to be scaled directly from the perceived familiarity of a probe. This assumption was tested in 2 experiments that examine the shape of confidence-based z receiver operating characteristic (zROC) curves under different levels of response bias induced by changing stimulus probabilities (Experiment 1) and payoffs (Experiment 2). Changes in the shape of the zROC curves with bias indicate that confidence is not scaled directly from perceived familiarity or likelihood. A model of information accumulation in recognition memory is proposed that can account for the observed effects.     

Similarity judgments and recognition memory for some common spices

A group of subjects rated the similarities of all 55 pairs of 11 spices, Nonmetric multidimensional scaling of this judgment matrix indicated a single underlying dimension, interpretable as pleasantness. Short-term recognition memory for the same spices was tested in separate groups of subjects required to remember one, three, and five stimuli, respectively. Performance was best for the group remembering one stimulus, but was not affected by the number of stimuli preceding the correct stimulus in the test. The confusion matrix for the one-stimulus group, symmetrized by removal of a stimulus bias, showed a significant correlation with the judgment matrix.     

Visual recognition memory for large and small binary pictures.

Recognition memory for pictures was tested using matrices of 0s and 1s as stimuli in two experiments. In the first experiment dimensionality of matrices was varied (5 X 5 and 10 X 10 matrices). The results showed better recognition memory for 10 X 10 matrices. The second experiment investigated the importance of certain areas, such as center or surround, using 10 X 10 matrices only. The results suggested that neither center nor surround is given unusual weight.     

Recognition memory for faces: When familiarity supports associative recognition judgments

Recognition memory for single items can be dissociated from recognition memory for the associations between items. For example, recognition tests for single words produce curvilinear receiver operating characteristics (ROCs), but associative recognition tests for word pairs produce linear ROCs. These dissociations are consistent with dual-process theories of recognition and suggest that associative recognition relies on recollection but that item recognition relies on a combination of recollection and assessments of familiarity. In the present study, we examined associative recognition ROCs for facial stimuli by manipulating the central and external features, in order to determine whether linear ROCs would be observed for stimuli other than arbitrary word pairs. When the faces were presented upright, familiarity estimates were significantly above zero, and the associative ROCs were curvilinear, suggesting that familiarity contributed to associative judgments. However, presenting the faces upside down effectively eliminated the contribution of familiarity to associative recognition, and the ROCs were linear. The results suggest that familiarity can support associative recognition judgments, if the associated components are encoded as a coherent gestalt, as in upright faces.     

Instability of individual differences in the association between confidence judgments and memory performance

There are large individual differences in the degree of association between the accuracy of memories and subjective confidence in those memories. Are these differences stable within the same test, and between alternate forms of a test? In Experiment 1, college students were tested on 3 recognition memory tasks, then retested 2 weeks later on alternate forms of the same tasks. The relationship between confidence judgments and recognition performance displayed low split-half stability and low alternate-forms stability. A second experiment with elderly adults replicated these findings. In a third experiment, college students recalled answers to general knowledge questions and rated confidence in the correctness of each answer. Individual differences in the association between confidence and recall performance were not stable across the odd- and even-numbered items on the test. These data indicate the need for the development of procedures that will produce stable estimates of individuals’ metacognitive accuracy.     

The role of the decision system in sensory and memory experiments using confidence judgments

This paper uses an artificial “sensory task” (a dice game) to investigate whether it is possible to distinguish between two-state and multistate sensory processes by examining confidence ratings of a group of unpracticed Os. The results indicate that Os use a decision rule that is deterministic so that both operating characteristics and a posteriori probability functions reflect the number of states used by the decision system. Thus, it would seem to be possible to use confidence ratings of Os to evaluate theoretical issues.     

Evidence for perceptual learning in pigeons' recognition memory for pictures

In two experiments, pigeons were trained on a recognition memory task, which required them to refrain from responding to a picture seen earlier that day. They learned this discrimination without detectable reliance on cues relating to the sequence of positive and negative trials. In both experiments, performance was significantly better when the same restricted set of stimuli was used each day than when an entirely novel set of stimuli was used each day, and in the former case there was less evidence of any significant decline in performance with increases in the interval between first and second presentations of a stimulus. The results suggest a powerful perceptual learning effect.     

Priors and payoffs in confidence judgments

Attention, Perception, & Psychophysics - Priors and payoffs are known to affect perceptual decision-making, but little is understood about how they influence confidence judgments. For optimal...     

Local and global judgments of confidence

Studies of calibration have shown that people's mean confidence in their answers (local confidence) tends to be greater than their overall estimate of the percentage of correct answers (global confidence). Moreover, whereas the former exhibits overconfidence, the latter often exhibits underconfidence. Three studies present evidence that global underconfidence reflects a failure to make an allowance for correct answers that are likely to result from mere guessing and can be eliminated by informing participants of the dubious normative status of estimates below 50% (i.e., chance). Previously reported discrepancies between global and local confidence, it is concluded, arise less from possible methodological artifacts in assessment of local confidence than from normatively inappropriate assessments of global confidence.     

Repetition and laterality effects on recognition memory for words and pictures*

Recognition memory for a list of words was tested by presenting a series of items with Ss instructed to make positive responses to targets (list items) and negative responses to distractors (nonlist items). The test items were either words or pictures, and they were presented tachistoscopically either to the left or right visual field. The results showed mean response latencies to be generally faster for stimuli presented to the right visual field. Response times were faster for target and distractor stimuli on their second test presentations than on initial tests, but this effect was much larger for targets. Repetitions were shown to decrease the amount of time necessary to execute the stimulus encoding and initial retrieval stages of recognition. This was also true, although to a lesser extent, if different stimulus forms (words or pictures) were used on the two tests. Subsequent recognition stages, including memory search and decision processes. were apparently independent of test stimulus form.     

Long-term perceptual specificity effects in recognition memory: the transformed pictures paradigm.

The effects of a study/test mismatch in the viewing mode of natural scenes on recognition memory performance were examined. At both encoding and retrieval, scenes were presented either by being divided into quarters that were displayed in a sequential cumulative fashion or by scrolling the images through the screen, thereby gradually revealing the content of the images. Half of the participants were tested immediately after encoding and the other half after 48 hours. For both the immediate and delayed retrieval conditions, better recognition memory was demonstrated when viewing modes matched across study and test than when they mismatched. Implications for current processing and multiple systems views of memory are discussed.     

Directed forgetting for categorised pictures: recognition memory for perceptual details versus gist

ABSTRACT

Item-based directed forgetting (DF) was tested using 2-alternative forced-choice recognition to examine the effects of forgetting instructions on memory for perceptual detail and gist of categorised pictures of scenes and objects in three experiments. When the distractor is from the same category as the target (exemplar test condition), discrimination must be based on memory for perceptual details, whereas recognition can be based on gist or general category information when the distractor is from a novel category (novel test condition). Recognition accuracy was greater for Remember-cued than Forget-cued pictures when discrimination must be based on perceptual details in the exemplar test condition but not when discrimination could be based on gist in the novel test condition. Accuracy in scene and object recognition is greater for gist than for perceptual details, and DF instructions serve to reduce recognition memory based on perceptual details.