Acquisition and retention in short-term memory

Serial attention to serial memory: The psychological refractory period in forward and backward cued recall

Guided by the conjecture that memory retrieval is attention turned inward, we examined serial attention in serial memory, combining the psychological refractory period (PRP) procedure from attention research with cued recall of two items from brief six-item lists. We report six experiments showing robust PRP effects in cued recall from memory (1-4) and cued report from perceptual displays (5-6), which suggest that memory retrieval requires the same attentional bottleneck as "retrieval" from perception. There were strong direction effects in each memory experiment. Response time (RT) was shorter and accuracy was higher when the cues occurred in the forward direction (left-to-right, top-to-bottom, first-to-last), replicating differences between forward and backward serial recall. Cue positions had strong effects on RT and accuracy in the memory experiments (1-4). The pattern suggested that subjects find cued items in memory by stepping through the list from the beginning or the end, with a preference for starting at the beginning. The perceptual experiments (5-6) showed weak effects of position that were more consistent with direct access. In all experiments, the distance between the cues in the list (lag) had weak effects, suggesting that subjects searched for each cue from the beginning or end of the list more often than they moved through the list from the first cue to the second. Direction, distance, and lag effects on RT and inter-response interval changed with SOA in a manner that suggested they affect bottleneck or pre-bottleneck processes that create and execute a plan for successive retrievals. We conclude that sequential retrieval from memory and sequential attention to perception engage the same computations and we show how computational models of memory can be interpreted as models of attention focused on memory.

Temporal weights in the perception of sound intensity: Effects of sound duration and number of temporal segments

Loudness is a fundamental aspect of auditory perception that is closely related to the physical level of the sound. However, it has been demonstrated that, in contrast to a sound level meter, human listeners do not weight all temporal segments of a sound equally. Instead, the beginning of a sound is more important for loudness estimation than later temporal portions. The present study investigates the mechanism underlying this primacy effect by varying the number of equal-duration temporal segments (5 and 20) and the total duration of the sound (1.0 to 10.0 s) in a factorial design. Pronounced primacy effects were observed for all 20-segment sounds. The temporal weights for the five-segment sounds are similar to those for the 20-segment sounds when the weights of the segments covering the same temporal range as a segment of the five-segment sounds are averaged. The primacy effect can be described by an exponential decay function with a time constant of about 200 ms. Thus, the temporal weight assigned to a specific temporal portion of a sound is determined by the time delay between sound onset and segment onset rather than by the number of segments or the total duration of the sound.

The Effect of Presentation Rate on Categorization

Homogeneous experimental (acoustically and semantically related) and control lists were presented to different groups of subjects at either 1 s or 6 s per word. The difference score between these two lists was used as a measure of categorization. Only words in the middle serial positions (representing retrieval from secondary memory) were scored. Clustering was also measured.

Slow presentation rate significantly increased categorization of the semantically similar words. This was not true of acoustically similar words. A possible explanation of this is that semantic categorization is time-consuming and acoustic categorization is not. Alternatively, the acoustic similarity effect may be a retrieval effect. The absence of acoustic clustering would seem to fit in with this latter interpretation.

Two Experiments with Artificial Languages

Two experiments are reported where subjects learnt simple artificial languages which contained synonyms: i.e. there were two equally correct verbal responses paired with each “referent” in the language: “referents” were simple visual patterns. In the first experiment (a memory task) subjects performed better with material of low auditory discriminability than with material of high auditory discriminability when they were asked to use one synonym of each pair more often than the other; the reverse was the case when they were asked to use each synonym equally often. In the second experiment (a two-person communication task with a noisy channel) the degree to which pairs of subjects tended to use the same synonym for the same meaning was found to be non-monotonic with the noise level in the communication channel: subjects tended to use the same synonyms for moderately high noise and very low noise more often than they did for very high noise and moderately low noise.

The concept of decision cost (Smith, 1968) is used to explain these results. Decision cost is a complex function both of discriminability and of response bias, and it is proposed that changes in decision cost in response to changes in response bias are responsible for the results observed in these experiments.

Absence of a Cross-Modal “Suffix Effect” in Short-Term Memory

Three experiments are reported involving the presentation of lists of either letters or digits for immediate serial recall. The main variable was the presence or absence of a suffix-prefix, an item (tick or cross) occurring at the end of the list which had to be copied before recall of the stimulus list. With auditory stimuli and an auditory suffix-prefix there was a large and selective increase in the number of errors on the last few serial positions—the typical “suffix effect”. The suffix effect was not found with auditory stimuli and a visual suffix-prefix nor with a visual stimulus and an auditory suffix-prefix. These results are interpreted as supporting a model for short-term memory proposed by Crowder and Morton (1969) in which it is suggested that with serial recall information concerning the final items following auditory presentation has a different, precategorical, origin from that concerning other items.

Retrieval Time and Decay of Information in Dichotic Memory

Self-paced recall in Broadbent's split-span memory task shows particular temporal patterns associated with different strategies of report. Accuracy is a decreasing function of the interval between presentation of an item and report of that item irrespective of report strategy. Presentation rate appears to affect the asymptote of this function but not its rate of decline. Results are related to an interaction between strategies of response selection and decay of information from memory.

A Unifying Principle of Memory Span, Forgetting, and Serial Position Effect

Lists of 5 to 10 digits, spoken at the rate of 2 digits per second, were presented to 10 Ss. The last digit in each list also occurred earlier in the list, and S's task was to write down the digit that followed the last one on its prior occurrence. The data were analyzed as measures of memory span, forgetting, and the serial position effect. These three phenomena were then related to one another by means of a principle that described how the probability of recall of each item in a list varied as a function of the numbers of preceding and following items. The effects of presentation rate and delayed recall were also discussed.

Detection Theory and Memory Processes: Are They Compatible?

Recent retention research has applied the concepts of the Tanner and Swets signal detection model to recognition and recall measures. While the literature is not extensive, the evidence tentatively suggests that d' and criterion changes are potentially useful descriptive constructs. Several methodological problems were noted in the literature reviewed. Specifically, comparative analyses of the data are difficult since different detection procedures were employed to estimate d' and criterion changes as a function of the same retention variables. In addition, an empirical relationship needs to be established between memory system output (e.g., response strength, familiarity, etc.) and the rating scale values which are fundamental to a retention detection analysis. Finally, a clear framework is needed for d' and criterion changes in memory.

Memory impairment caused by cerebral hematoma in the left medial temporal lobe due to ruptured posterior cerebral artery aneurysm

BACKGROUND

Cognitive disorders, such as memory disturbances, are often observed following a subarachnoid hemorrhage. We present a very rare case where rupture of a posterior cerebral artery aneurysm caused restricted damage to the hippocampus unilaterally, and caused memory disturbances.

CASE PRESENTATION

A 56-year-old, right-handed man, with a formal education history of 16 years and company employees was admitted to our hospital because of a consciousness disturbance. He was diagnosed as having a subarachnoid hemorrhage due to a left posterior cerebral artery dissecting aneurysm, and coil embolization was performed. Subsequently, he had neither motor paresis nor sensory disturbances, but he showed disorientation, and both retrograde and anterograde amnesia. Although immediate recall and remote memory were almost intact, his recent memory was moderately impaired. Both verbal and non-verbal memories were impaired. Brain computed tomography (CT) and magnetic resonance imaging (MRI) revealed a cerebral hematoma in the left temporal lobe involving the hippocampus and parahippocampal gyrus, and single-photon emission computed tomography (SPECT) demonstrated low perfusion areas in the left medial temporal lobe.

CONCLUSIONS

We suggest that the memory impairment was caused by local tissue destruction of Papez's circuit in the dominant hemisphere due to the cerebral hematoma.

[Temporal limitations of visual attention for order perception]

The relative order of a letter sequence is difficult to recognize when it is presented repeatedly than when it is presented only once in a rapid serial visual presentation (Holcombe, Kanwisher, & Triesman, 2001). In the present study, we investigated a critical factor that causes this order deficit. Experiment 1 demonstrated that repeating a letter sequence in a short time period induced the order deficit. The robust order deficit did not recover even when the letter sequence included a salient item that attracted exogenous attention. Experiment 2 showed that attending to a briefly presented letter triggered the order deficit for a letter sequence presented within 500 ms after the preceding letter. These results suggest that the order deficit is caused by the temporal limits of visual attention that prevent order information from being consolidated into visual working memory.

Toward a theory of human memory: Data structures and access processes

Starting from Marr's ideas about levels of explanation, a theory of the data structures and access processes in human memory is demonstrated on 10 tasks. Functional characteristics of human memory are captured implementation-independently. Our theory generates a multidimensional task classification subsuming existing classifications such as the distinction between tasks that are implicit versus explicit, data driven versus conceptually driven, and simple associative (two-way bindings) versus higher order (threeway bindings), providing a broad basis for new experiments. The formal language clarifies the binding problem in episodic memory, the role of input pathways in both episodic and semantic (lexical) memory, the importance of the input set in episodic memory, and the ubiquitous calculation of an intersection in theories of episodic and lexical access.

Function transformation without reinforcement

In studies of function transformation, participants initially are taught to match stimuli in the presence of a contextual cue, X; the stimuli to be matched bear some formal relation to each other, for example, a relation of opposition or difference. In a second phase, the participants are taught to match arbitrary stimuli (say, A and B) in the presence of X. In a final test, A often displays behavioral functions that differ from those of B, and can be predicted from the nature of the relation associated with X in the initial training phase. Here we report function-transformation effects in the absence of selection responses and of their reinforcers. In three experiments with college students, exposure to relations of difference or identity modified the responses given to later stimuli. In Experiment 1, responses to a test stimulus A varied depending on preexposure to pairs of colors that were distinct from A but exemplified relations of difference or identity. In Experiment 2, a stimulus A acquired distinct functions, depending on its previous pairing with a contextual cue X that had itself been paired with identity or difference among colors. Experiment 3 confirmed the results of Experiment 2 with a modified design. Our data are consistent with the notion that relations of identity or difference can serve as stimuli for Pavlovian processes, and, in compound with other cues, produce apparent function-transformation effects.

Why is it difficult to see in the fog? How stimulus contrast affects visual perception and visual memory

Processing visually degraded stimuli is a common experience. We struggle to find house keys on dim front porches, to decipher slides projected in overly bright seminar rooms, and to read 10th-generation photocopies. In this research, we focus specifically on stimuli that are degraded via reduction of stimulus contrast and address two questions. First, why is it difficult to process low-contrast, as compared with high-contrast, stimuli? Second, is the effect of contrast fundamental in that its effect is independent of the stimulus being processed and the reason for processing the stimulus? We formally address and answer these questions within the context of a series of nested theories, each providing a successively stronger definition of what it means for contrast to affect perception and memory. To evaluate the theories, we carried out six experiments. Experiments 1 and 2 involved simple stimuli (randomly generated forms and digit strings), whereas Experiments 3-6 involved naturalistic pictures (faces, houses, and cityscapes). The stimuli were presented at two contrast levels and at varying exposure durations. The data from all the experiments allow the conclusion that some function of stimulus contrast combines multiplicatively with stimulus duration at a stage prior to that at which the nature of the stimulus and the reason for processing it are determined, and it is the result of this multiplicative combination that determines eventual memory performance. We describe a stronger version of this theory--the sensory response, information acquisition theory--which has at its core, the strong Bloch's-law-like assumption of a fundamental visual system response that is proportional to the product of stimulus contrast and stimulus duration. This theory was, as it has been in the past, highly successful in accounting for memory for simple stimuli shown at short (i.e., shorter than an eye fixation) durations. However, it was less successful in accounting for data from short-duration naturalistic pictures and was entirely unsuccessful in accounting for data from naturalistic pictures shown at longer durations. We discuss (1) processing differences between short- and long-duration stimuli, (2) processing differences between simple stimuli, such as digits, and complex stimuli, such as pictures, (3) processing differences between biluminant stimuli (such as line drawings with only two luminance levels) and multiluminant stimuli (such as grayscale pictures with multiple luminance levels), and (4) Bloch's law and a proposed generalization of the concept of metamers.

Writing and overwriting short-term memory

An integrative account of short-term memory is based on data from pigeons trained to report the majority color in a sequence of lights. Performance showed strong recency effects, was invariant over changes in the interstimulus interval, and improved with increases in the intertrial interval. A compound model of binomial variance around geometrically decreasing memory described the data; a logit transformation rendered it isomorphic with other memory models. The model was generalized for variance in the parameters, where it was shown that averaging exponential and power functions from individuals or items with different decay rates generates new functions that are hyperbolic in time and in log time, respectively. The compound model provides a unified treatment of both the accrual and the dissipation of memory and is consistent with data from various experiments, including the choose-short bias in delayed recall, multielement stimuli, and Rubin and Wenzel's (1996) meta-analyses of forgetting.

Effect of sensory modality for presentation and word imagery on paired-associate learning

Recall performance on a paired-associated learning task was investigated as a function of word imagery modality (visual or auditory), presentation mode (visual or auditory), and sex. Analysis showed greater recall of visual imagery words, and the results are consistent with Paivio's (1971) conceptual-peg hypothesis. Visual presentation of word lists produced greater recall than auditory presentation, and females exhibited greater recall performance than did males. A predicted interaction between modality for presentation and for word imagery did not reach statistical significance. The implications for future research with sensory imagery in learning is discussed.

Forgetting: trace erosion or retrieval failure?

A series of lists of random words was presented. Following each list, the subject attempted to recall the words of the list prior to the list just presented. Recall probability for a given word depended on the length of the list in which it was embedded, not on the length of the list intervening between presentation and test. These results indicate that forgetting is a failure in the memory search during retrieval rather than a degradation of the memory trace occurring between presentation and test.

Independent functioning of verbal memory stores: a neuropsychological study

Five experiments are described concerning verbal short-term memory performance of a patient who has a very markedly reduced verbal span. The results of the first three, free recall, the Peterson procedure and an investigation of proactive interference, indicate that he has a greatly reduced short-term memory capacity, while the last two, probe recognition and missing scan, show that this cannot be attributed to a retrieval failure. Since his performance on long-term memory tasks is normal, it is difficult to explain these results with theories of normal functioning in which verbal STM and LTM use the same structures in different ways. They also make the serial model of the relation between STM and LTM less plausible and support a model in which verbal STM and LTM have parallel inputs.