Study repetition and the rejection of conjunction lures

The constructive nature of memories in insects: bumblebees as a case study

The view that human memory is constructive implies that recollections are not necessarily an accurate reproduction of past events. An approach to study this constructive nature of memory is by examining memory errors. In this regard, conjunction errors-i.e. incorrect recollection of new stimuli integrated by components from two previously studied stimuli-have attracted important attention in human memory research. Do animals other than humans make conjunction errors? To investigate this issue, a choice task in which training was not involved was used. Bees experienced two to-be-remembered stimuli. At the test, they were presented with four stimuli: one of the original items (i.e. old), an item made by combining two features of the original items (i.e. conjunction), an item containing a previously presented feature and a new one (i.e. feature), and an item integrated solely by new features (i.e. new). Bumblebees remembered the old items. Importantly, when making memory errors, bumblebees selected conjunction and feature lures more often than new items. These results indicate that bumblebees, like humans, spontaneously make memory conjunction errors and suggest that invertebrates' memories might also be constructive in nature. I suggest that focusing on memory errors is a solid avenue to investigate episodic (like) memory in animals.This article is part of the theme issue 'Elements of episodic memory: lessons from 40 years of research'.

Learning facilitates dual-process face recognition regardless of holistic processing

Much evidence suggests that faces are recognized based on their global familiarity in a signal-detection manner. However, experiments drawing this conclusion typically present study lists of faces only once or twice, and the nature of face recognition at higher levels of learning remains unclear. Here, three experiments are reported in which participants studied some faces eight times and others twice and then took a recognition test containing previously viewed faces, entirely new faces, and faces which recombined the parts of previously viewed faces. Three measures converged to suggest that study list repetition increased the likelihood of participants rejecting recombined faces as new by recollecting that their parts were studied but in a different combination, and that manipulating holistic or Gestalt-like processing-a hallmark of face perception-qualitatively preserved its effect on how memory judgments are made. This suggests that face learning causes a shift from the use of a signal-detection strategy to the use of a dual-process strategy of face recognition regardless of holistic processing.

Facial Recall: Feature–Conjunction Effects in Source Retrieval Versus Item Recognition

Within memory processing, feature and conjunction effects refer to higher false alarms (incorrectly naming items as previously seen) for stimuli with partial/all features taken from different previously studied versus novel items. In this study, we compared feature–conjunction effects in the memory tasks of both item recognition and source retrieval, using faces as stimuli. We found greater feature–conjunction effects on source retrieval versus item recognition tasks, though participants gave slower responses on source retrieval versus item recognition. Thus, our data were consistent with dual-process memory theory in which familiarity contributes to high false alarms in compound faces even while feature–conjunction effects enhance recollection in source retrieval.

Dissociating the Electrophysiological Correlates between Item Retrieval and Associative Retrieval in Associative Recognition: From the Perspective of Directed Forgetting

Although many behavioral studies have reported associative memory was different from item memory, evidence coming from ERP researches has been in debate. In addition, directed forgetting effect for items has been fully discussed, but whether association between items can be directed-forgotten was unclear. The directed forgetting effect was important for dissociating the item retrieval and associative retrieval because of the one-to-one mapping relationship both between item retrieval and familiarity and between associative retrieval and recollection. Thus, the aim of this study was to investigate the dissociation between item retrieval and associative retrieval and test directed forgetting effect for associative information. Associative recognition paradigm combined with directed forgetting paradigm by ERP recording was employed. Old/rearranged effect in to-be-remembered condition, which was associated with associative memory, was significant at 500–800 ms (LPC) but not at 300–500 ms interval (FN400), indicating that item information was retrieved prior to associative information. The ERP wave calculated by subtracting the to-be-forgotten old pairs with “old” response from those with “rearranged” response, which reflected associative retrieval in the to-be-forgotten condition, was negative from 500 to 800 ms (reversed old/new effect), indicating that association between items can be directed-forgotten. Similar evidence was obtained by contrasting “rearranged” responses aimed to the to-be-forgotten old pairs with those aimed to the to-be-remembered rearranged pairs, which actually represented the complete failure of associative retrieval. Therefore, item retrieval and associative retrieval were indexed by FN400 and LPC respectively, with associative retrieval more inhibited than item retrieval.

Distinguishing familiarity from fluency for the compound word pair effect in associative recognition

We examined whether processing fluency contributes to associative recognition of unitized pre-experimental associations. In Experiments 1A and 1B, we minimized perceptual fluency by presenting each word of pairs on separate screens at both study and test, yet the compound word (CW) effect (i.e., hit and false-alarm rates greater for CW pairs with no difference in discrimination) did not reduce. In Experiments 2A and 2B, conceptual fluency was examined by comparing transparent (e.g., hand bag) and opaque (e.g., rag time) CW pairs in lexical decision and associative recognition tasks. Lexical decision was faster for transparent CWs (Experiment 2A) but in associative recognition, the CW effect did not differ by CW pair type (Experiment 2B). In Experiments 3A and 3B, we examined whether priming that increases processing fluency would influence the CW effect. In Experiment 3A, CW and non-compound word pairs were preceded with matched and mismatched primes at test in an associative recognition task. In Experiment 3B, only transparent and opaque CW pairs were presented. Results showed that presenting matched versus mismatched primes at test did not influence the CW effect. The CW effect in yes-no associative recognition is due to reliance on enhanced familiarity of unitized CW pairs.

Effects of repetition on age differences in associative recognition

The study reported here examined the effect of repetition on age differences in associative recognition using a paradigm designed to encourage recollection at test. Young and older adults studied lists of unrelated word pairs presented one, two, four, or eight times. Test lists contained old (intact) pairs, pairs consisting of old words that had been studied with other partners (rearranged lures), and pairs consisting of two unstudied words (new lures). Participants gave old/new responses and then indicated whether their responses were based on details that they could recollect or on familiarity. Older adults exhibited an ironic effect of repetition-an increase in false alarms on rearranged lures with more study opportunities-whereas young adults did not. Older adults also claimed to recall details of the study episode for rearranged lures whose constituent words were presented more frequently, but this was not true for young adults. Although both young and older adults said that they based correct rejections of rearranged lures on memory for details of the study episode, this effect was stronger for young adults. The observed age differences are consistent with older adults having reduced use of recollection in associative recognition tasks.

The Neural Basis of Recollection Rejection: Increases in Hippocampal-Prefrontal Connectivity in the Absence of a Shared Recall-to-Reject and Target Recollection Network

Recollection rejection or "recall-to-reject" is a mechanism that has been posited to help maintain accurate memory by preventing the occurrence of false memories. Recollection rejection occurs when the presentation of a new item during recognition triggers recall of an associated target, a mismatch in features between the new and old items is registered, and the lure is correctly rejected. Critically, this characterization of recollection rejection involves a recall signal that is conceptually similar to recollection as elicited by a target. However, previous neuroimaging studies have not evaluated the extent to which recollection rejection and target recollection rely on a common neural signal but have instead focused on recollection rejection as a postretrieval monitoring process. This study utilized a false memory paradigm in conjunction with an adapted remember-know-new response paradigm that separated "new" responses based on recollection rejection from those that were based on a lack of familiarity with the item. This procedure allowed for parallel recollection rejection and target recollection contrasts to be computed. Results revealed that, contrary to predictions from theoretical and behavioral literature, there was virtually no evidence of a common retrieval mechanism supporting recollection rejection and target recollection. Instead of the typical target recollection network, recollection rejection recruited a network of lateral prefrontal and bilateral parietal regions that is consistent with the retrieval monitoring network identified in previous neuroimaging studies of recollection rejection. However, a functional connectivity analysis revealed a component of the frontoparietal rejection network that showed increased coupling with the right hippocampus during recollection rejection responses. As such, we demonstrate a possible link between PFC monitoring network and basic retrieval mechanisms within the hippocampus that was not revealed with univariate analyses alone.

The role of familiarity in associative recognition of unitized compound word pairs

This study examined the effect of unitization and contribution of familiarity in the recognition of word pairs. Compound words were presented as word pairs and were contrasted with noncompound word pairs in an associative recognition task. In Experiments 1 and 2, yes–no recognition hit and false-alarm rates were significantly higher for compound than for noncompound word pairs, with no difference in discrimination in both within- and between-subject comparisons. Experiment 2 also showed that item recognition was reduced for words from compound compared to noncompound word pairs, providing evidence of the unitization of the compound pairs. A two-alternative forced-choice test used in Experiments 3A and 3B provided evidence that the concordant effect for compound word pairs was largely due to familiarity. A discrimination advantage for compound word pairs was also seen in these experiments. Experiment 4A showed that a different pattern of results is seen when repeated noncompound word pairs are compared to compound word pairs. Experiment 4B showed that memory for the individual items of compound word pairs was impaired relative to items in repeated and nonrepeated noncompound word pairs, and Experiment 5 demonstrated that this effect is eliminated when the elements of compound word pairs are not unitized. The concordant pattern seen in yes–no recognition and the discrimination advantage in forced-choice recognition for compound relative to noncompound word pairs is due to greater reliance on familiarity at test when pairs are unitized.

Improving associative memory in older adults with unitization

We examined if unitization inherent preexperimentally could reduce the associative deficit in older adults. In Experiment 1, younger and older adults studied compound word (CW; e.g., store keeper) and noncompound word (NCW; e.g., needle birth) pairs. We found a reduction in the age-related associative deficit such that older but not younger adults showed a discrimination advantage for CW relative to NCW pairs on a yes-no associative recognition test. These results suggest that CW compared to NCW word pairs provide schematic support that older adults can use to improve their memory. In Experiment 2, reducing study time in younger adults decreased associative recognition performance, but did not produce a discrimination advantage for CW pairs. In Experiment 3, both older and younger adults showed a discrimination advantage for CW pairs on a two-alternative forced-choice recognition test, which encourages greater use of familiarity. These results suggest that test format influenced young adults' use of familiarity during associative recognition of unitized pairs, and that older adults rely more on familiarity than recollection for associative recognition. Unitization of preexperimental associations, as in CW pairs, can alleviate age-related associative deficits.

Need for cognition is related to the rejection (but not the acceptance) of false memories

Need for cognition (NFC) and rates of false recognition were studied in the context of the memory conjunction paradigm. Past research has shown that NFC is related to false recognition and false recall (Graham, 2007; Leding, 2011) in the Deese-Roediger-McDermott (DRM) paradigm, with high-NFC people experiencing more false memories. The present study extended this research to determine whether the findings of high-NFC people experiencing higher levels of false memories could be extended to other false memory paradigms. The present study also examined rates of high-confidence rejections of lures and recollection rejection responses. It was found that there were no significant differences for high- and low-NFC people in false recognition but that high-NFC people were more likely to use high-confidence rejections and have recollection rejection responses when they did reject lures.

Contributions of familiarity and recollection rejection to recognition: evidence from the time course of false recognition for semantic and conjunction lures

It has been suggested that both familiarity and recollection contribute to the recognition decision process. In this paper we leverage the form of false alarm rate functions--in which false alarm rates describe an inverted U-shaped function as the time between study and test increases--to assess how these processes support retention of semantic and surface form information from previously studied words. We directly compare the maxima of these functions for lures that are semantically related and lures that are related by surface form to previously studied material. This analysis reveals a more rapid loss of access to surface form than to semantic information. To separate the contributions of item familiarity and reminding-induced recollection rejection to this effect, we use a simple multinomial process model; this analysis reveals that this loss of access reflects both a more rapid loss of familiarity and lower rates of recollection for surface form information.

Recall-to-reject: The effect of category cues on false recognition

Four experiments examined the effect of category cueing on recall-to-reject, one of the central memory-editing mechanisms thought to prevent the occurrence of false memories. When category names were used as retrieval cues, the typically observed false recognition effect was eliminated for semantically associated distractors (Experiment 1a) and, moreover, a reduction in the absolute level of the false alarm rate was found for phonologically associated distractors (Experiment 2a). In addition to the old/new-recognition data, analyses using multinomial models support the interpretation that category cueing was successful in increasing the probability of recall-to-reject (Experiments 1b and 2b). The results are in line with dual-process theories of recognition memory and provide further evidence for recall-to-reject in single item recognition. They demonstrate its potential to reduce false recognition even when explicit instructions are not given. In addition, the results demonstrate that the paradigm can give rise to side effects that oppose recall-to-reject. A simultaneous familiarity increase can explain why many studies failed to find evidence for recall-to-reject in terms of false alarm rates.

Context-dependent repetition effects on recognition memory

One widely acknowledged way to improve our memory performance is to repeatedly study the to be learned material. One aspect that has received little attention in past research regards the context sensitivity of this repetition effect, that is whether the item is repeated within the same or within different contexts. The predictions of a neuro-computational model (O'Reilly & Norman, 2002) were tested in an experiment requiring participants to study visual objects either once or three times. Crucially, for half of the repeated objects the study context (encoding task, background color and screen position) remained the same (within context repetition) while for the other half the contextual features changed across repetitions (across context repetition). In addition to behavioral measures, event-related potentials (ERP) were recorded that provide complementary information on the underlying neural mechanisms during recognition. Consistent with dual-process models behavioral estimates (remember/know-procedure) demonstrate differential effects of context on memory performance, namely that recognition judgements were more dependent on familiarity when repetition occurs across contexts. In accordance with these behavioral results ERPs showed a larger early frontal old/new effect for across context repetitions as compared to within context repetitions and single presentations, i.e. an increase in familiarity following repetition across study contexts. In contrast, the late parietal old/new effect, indexing recollection did not differ between both repetition conditions. These results suggest that repetition differentially affects familiarity depending on whether it occurs within the same context or across different contexts.

Conjunction errors and semantic transparency

Memory conjunction errors occur when aspects of two different events are falsely recognized or recalled as having occurred as parts of the same event. One theoretical account of conjunction errors is rooted in traditional dual-process models of recognition judgments, in which responses are based on an item's familiarity or the retrieval of recollected details associated with the encoding of that item. We manipulated the familiarity of test probes by varying their semantic overlap with studied items, taking advantage of the inherent semantic transparency of compound words. Transparent compounds are those whose component parts (lexemes) are semantically related to the meaning of the entire word. In contrast, opaque compounds' lexemes do not contribute directly to the meaning of the compound. We showed that the familiarity of semantically transparent assembly lures created from their lexemes (study dog and house, test on doghouse) is greater than the familiarity of opaque assembly lures (study back and draw, test on drawback). A response-signal experiment revealed no evidence for the use of a recall-to-reject process for either semantically transparent or opaque lures.

Memory conjunction errors: the effects of presentation duration and study repetition

The memory-strengthening manipulations of increased presentation duration and increased number of times items were presented were manipulated in the memory conjunction paradigm. Participants viewed parent words once or three times during the study portion of the experiment for 250 ms, 1000 ms, or 3000 ms. After an old/new recognition test participants were asked to give explanations for their answers from the recognition test. The results of true and false recognition as well as recall-to-reject responses (e.g., I know I did not see blackbird since I saw blackmail) indicated that both familiarity and recollection were influenced by the memory-strengthening manipulations. The results provide evidence for dual-process theories of recognition memory and the opposing processes of familiarity and recollection.

When false recognition is out of control: the case of facial conjunctions

In three experiments, a dual-process approach to face recognition memory is examined, with a specific focus on the idea that a recollection process can be used to retrieve configural information of a studied face. Subjects could avoid, with confidence, a recognition error to conjunction lure faces (each a reconfiguration of features from separate studied faces) or feature lure faces (each based on a set of old features and a set of new features) by recalling a studied configuration. In Experiment 1, study repetition (one vs. eight presentations) was manipulated, and in Experiments 2 and 3, retention interval over a short number of trials (0-20) was manipulated. Different measures converged on the conclusion that subjects were unable to use a recollection process to retrieve configural information in an effort to temper recognition errors for conjunction or feature lure faces. A single process, familiarity, appears to be the sole process underlying recognition of conjunction and feature faces, and familiarity contributes, perhaps in whole, to discrimination of old from conjunction faces.

A decrease in conjunction error rates across lags on a continuous recognition task: A robust pattern

In four experiments, the lag retention interval from parent words (e.g., blackmail, jailbird) to a conjunction word ( blackbird) was manipulated in a continuous recognition task. Alterations to the basic procedure of Jones and Atchley (2002) were employed in Experiments 1 and 2 to bolster recollection to reject conjunction lures, yet conjunction error rates still decreased across lags of 1 to 20 words. Experiment 3 and a multiexperiment analysis examined the increments of forgetting in familiarity across lags of 1–20 words. Finally, in Experiment 4, participants attempted to identify conjunction probes as “old”, and the data were contrasted with those from a previous experiment (Jones & Atchley, 2002, Exp. 1), in which participants attempted not to identify conjunction probes as “old”. In support of earlier findings, the decrease in familiarity across lags of 1–20 words appears robust, with a constant level of weak recollection occurring for parent words.

Metamemorial influences in recognition memory: pictorial encoding reduces conjunction errors

Two experiments are presented that explore the role of the distinctiveness heuristic (e.g., Schacter, Israel, & Racine, 1999) on rates of conjunction errors as a function of encoding condition. The results of Experiment 1 demonstrate a reliable reduction of conjunction errors when participants study pictures relative to both reading words aloud and silently. Experiment 2 demonstrates that the nature of the pictures presented during the study phase is important for reducing conjunction errors. Collectively, the experiments demonstrate that participants can use the distinctiveness heuristic in addition to recall-to-reject strategies to avoid conjunction errors. These findings add to a growing body of literature that suggests that participants are able to use expectations for memory to guide their recognition decisions.

Feature and conjunction effects in recognition memory: Toward specifying familiarity for compound words

In three experiments, we evaluated potential sources of familiarity in the production of feature and conjunction errors with both word (Experiments 1 and 3) and nonword (Experiment 2) stimuli and related this work to various dual-process models of memory. The contributions of letter, syllable, lexical morpheme, and conceptual information were considered. Lexical morpheme information was consistently more potent than syllable information in leading to feature and conjunction errors across Experiments 1 and 2, and a word length explanation did not account for this consistent finding. In addition, there was no impact of conceptual information on these errors (Experiments 1-3). The results support a familiarity-based interpretation of feature and conjunction errors and a lexical morpheme basis for the familiarity in compound words. In order to be more comprehensive, memory models may need to account for a lexical morpheme source of familiarity.

Familiarity from orthographic information: Extensions of the recognition without identification effect

Two experiments were conducted to investigate the effect of letter location information in recognition memory judgments. The experiments used the recognition without identification paradigm (Peynircioglu, 1990), in which participants first attempt to identify the test item and then make a recognition decision as to whether or not the item was studied. In these studies, items that are not identified but that correspond to items that were presented are typically still rated as more likely to have been studied than those that were not presented. The present experiments demonstrated this finding with a variant of the conjunction lure paradigm. In Experiments 1 and 2, participants were tested with word fragments that were made from the letters of two words. When the letters were from studied items, fragments were rated higher than when the test items were derived from two unstudied items, or one studied item and one unstudied item, suggesting that recognition without identification is prone to the same types of errors as recognition with identification. Results are discussed in terms of familiarity effects in recognition memory.

Editing (out) generated study words in a recognition exclusion task: Effects of response signal delay and generation procedure

Three experiments investigated participants' ability to exclude generated study words from "old" judgements on an exclusion task in which a response signal delay (RSD) was manipulated. In Experiment 1, words were generated by mentally conjoining components of two compound words, and a conjunction test condition was included (e.g., presentation of the lure checkpoint after studying checklist and needlepoint without generating the target). The exclusion error rate for generated words was higher than the conjunction error rate for a short RSD group but lower than the conjunction error rate for a long RSD group. The decrease in the familiarity effect across RSDs was extended to a conceptual cue generation procedure in Experiments 2 and 3, but an anagram generation procedure did not show similar evidence of recollection-based exclusions. The earliest indication of recollection-based editing of generated words occurred in a 2250-ms RSD (Experiment 3). An extension of a multi-factor transfer appropriate processing theory (deWinstanley, Bjork, & Bjork, 1996) of generation effects that incorporates influences of familiarity and recollection is advocated.

Conjunction errors, recollection-based rejections, and forgetting in a continuous recognition task

Six experiments investigated conjunction memory errors (e.g., falsely remembering blackbird after studying parent words blackmail and jailbird) in a continuous recognition procedure with a parent-conjunction lag manipulation. In 4 experiments (1A, 1B, 2A, and 2B) "recollect" judgments, which indexed recall of parent words, showed that participants can use recollection to prevent conjunction errors. "Recollect" judgments, as well as overt recall of parent words (in Experiments 2A and 2B), dropped sharply from a lag of 0 to 1 word, then stabilized from a lag of 1 to 20 words. Thus, the "recollect" responses and overt recall demonstrate a step function of forgetting over short intervals. These data generalized to cued recall in Experiments 3A and 3B with the first morpheme (e.g., black) as the cue, though recall conjunction errors occurred infrequently relative to recognition conjunction errors. Overall, the results support the idea that automatic and controlled processes contribute to memory performance.

Conjunction errors in recognition memory: modality-free errors for older adults but not for young adults

A dual-process theory of memory was applied to processes in normal aging, with a focus on recognition errors in the feature-conjunction paradigm (i.e., false recognition of blackbird after studying parent words blackmail and/or jailbird). Study repetition was manipulated so that some parent words occurred once and others occurred three times. Age-related differences on hit scores occurred for two experiments. The results for feature and conjunction conditions showed repetition effects but no age-related differences when participants were uninformed of the lures (Experiment 1). However, age-related differences emerged when the retrieval of modality source information created a way to evade conjunction errors (Experiment 2). In the second experiment, study repetition decreased errors for the young adults but increased errors for the older adults, and young adults were better able than older adults to avoid conjunction errors when the parent words had been repeated. For older adults, the conjunction errors were modality-free. The results provide additional evidence that older adults experience difficulty in recollecting aspects of a study experience, and the results from groups of young adults required to respond quickly on the tests provide converging evidence for this conclusion.