An empirical note on the role of verbal labels in motor short-term memory tasks

Previous studies of motor short-term memory have shown that when a criterion movement on a semicircular positioning task is accompanied by an appropriate verbal label (a clock-face position), recall of the movement is more accurate than when only the movement is presented. This increased accuracy could be due to either the additional spatial information provided by the label or enhanced retention of the movement information. These two alternatives cannot be distinguished on the basis of previous studies because the studies have not evaluated movement accuracy following presentation of the label alone. The present study employed such a condition in addition to the movement-only and movement-plus-label conditions to distinguish between the two hypotheses. In all conditions, subjects were asked to move to the criterion position after a retention interval of either 5 sec or 60 sec. Evidence indicated that subjects who received both the label and the movement tended to use the spatial information provided by the label at the 60-sec interval. The evidence did not indicate that the verbal label actually enhanced retention of the movement information.     

Interference Between Location and Distance Information in Motor Short-Term Memory: The Respective Roles of Direct Kinesthetic Signals and Abstract Codes

Interference between location and distance information in motor short-term memory has been hypothesized on the basis of the systematic pattern of undershooting and overshooting in movement reproduction that occurs when the starting position for reproduction movements is shifted. To determine the possible contribution of limb-specific kinesthetic information to this systematic undershooting-overshooting pattern, we compared the reproduction of linear arm positioning movements performed under either same-limb or switched-limb conditions. Ten subjects were assigned to either a location or distance cue condition, and each subject completed a total of 40 trials, 20 under same-limb and 20 under switched-limb conditions. Each trial consisted of criterion and reproduction movements, separated by a 10-s retention interval. The starting position for the reproduction movement was shifted by 0, 2, or 4 cm in either direction from that of the criterion movement. The systematic undershooting-overshooting pattern, which occurs when either the movement location or distance is reproduced, arose under both the same-limb and switched-limb conditions, suggesting that the primary cause of the location-distance interference is not limb-specific kinesthetic information. Rather, more abstract information in the form of a conceptual memory code appears to be the probable cause of the location distance interference phenomenon.     

Dissociating cognitive and motor interference effects on kinesthetic short-term memory

In two experiments, we investigated how short-term memory of kinesthetically defined spatial locations suffers from either motor or cognitive distraction. In Exp. 1, 22 blindfolded participants moved a handle with their right hand towards a mechanical stop and back to the start and then reproduced the encoded stop position by a second movement. The retention interval was adjusted to approximately 0 and 8 s. In half of the trials participants had to provide a verbal judgment of the target distance after encoding (cognitive distractor). Analyses of constant and variable errors indicated that the verbal judgments interfered with the motor reproduction only, when the retention interval was long. In Exp. 2, 22 other participants performed the same task but instead of providing verbal distance estimations they performed an additional movement either with their right or left hand during the retention interval. Constant error was affected by the side of the interpolated movement (right vs. left hand) and by the delay interval. The results show that reproduction of kinesthetically encoded spatial locations is affected differently in long- and short-retention intervals by cognitive and motor interference. This suggests that reproduction behavior is based on distinct codes during immediate vs. delayed recall.     

Memory for Constrained and Preselected Movement Location and Distance

These experiments assessed the interrelationship between location and distance cues in the coding of movements. In separate experiments subjects recalled either the terminal location or the distance of constrained (Experiment 1) or preselected (Experiment 2) movements following a 15-sec retention interval. Changes in direction amd amplitude of starting position were used to ascertain whether recall errors were related to these changes. The findings of both experiments indicated that location and distance were recalled with similar accuracy when the starting position was identical for the criterion and recall movement. However, analysis of constant errors when the recall starting position was varied in either direction clearly indicated neither terminal location nor distance are coded independently, and memory for movement is based on an interaction between these cues.     

Visual vs. verbal stimulus set in a letter-classification task

In three experiments, subjects classified briefly presented letters as belonging to either the first or the second half of the alphabet. Prior to each target letter, the subjects were given either a verbally named letter (verbal set), a letter presented visually for 3 sec (visual set), or no prior alternative (control). The target was equally likely to be the same as the prior alternative (same trials) or from the opposite half of the alphabet (different trials). Classification accuracy was always greater for visual set than for the control condition. Verbal-set accuracy was no better than control accuracy when the verbal alternative immediately preceded the target but was equal to visual-set accuracy when the alternative preceded the target by 3 sec. In both set conditions, subjects tended to choose the same half of the alphabet as the prior alternative. It was concluded that type and timing of prior alternatives are important variables in accounting for enhanced classification accuracy.     

Effect of interpolated temporal activity on temporal short-term memory

The purpose of this research was to determine the retroactive interference effects of a single interpolated task (i.e., one temporal duration) on the retention of a criterion duration. This research is of interest because the mnemonic structure of temporal information of different durations is uncertain. Previous research has indicated that there might be a difference in structure for durations of 1 and 4 sec., although the results are inconsistent. Thus, two criterion durations of 1 and 4 sec. and five interpolated durations (i.e., 60%, 80%, 100%, 120%, and 140% of the duration of the criterion) were utilized under the method of reproduction. In addition, subjects were instructed to use either a counting strategy or none (referred to as conscious time estimation) to facilitate the retention of the temporal information. Recall was less variable when using a counting strategy than not and when estimating 1 sec. than 4 sec. However, there was no effect of interpolated activity when comparing performance across different interpolated conditions (no interpolated activity). Apparently, one interpolated duration is not sufficient to produce structural interference with a single criterion duration.     

Simultaneous processing of verbal and spatial information

Two experiments investigated the generality of the selective interference phenomenon by studying several different concurrent tasks, each involving either verbal or spatial information, and a range of retention intervals from 10 to 30 sec. The results supported the selective interference predictions, since the interference in performance was greater when the two simultaneous tasks both involved either verbal information or spatial information than when one of the tasks involved verbal and the other spatial information. It was concluded that not only was the selective interference phenomenon a rather general effect but that, since it was unaffected by changes in the nature of the component tasks, it was also probably caused by central processing capacity limitations rather than by competition for a specific structural mechanism.     

Using selective interference to investigate spatial memory representations

Two experiments used a selective interference procedure in an attempt to determine whether nonverbal visual stimuli were represented in memory in a verbal or spatial format. A spatial representation was clearly implicated. In both experiments, Ss were required to remember either the positions or the identities of seven target items in a 25-item array. During the retention interval for that information, Ss attempted to recognize schematic face or airplane photograph stimuli in a same-different memory task. Memory performance on one or both tasks was greatly impaired when the recall task involved position or spatial information, but was either much less or not at all affected by an identity or verbal information recall task. Because of the selective nature of the interference and on the basis of certain correlational evidence, the experimental results were also interpreted as providing support for the notion that verbal and spatial information are stored and processed in separate information-processing systems.     

Imagery ability in recall of distance and location information

The role of visual imagery in short-term retention of movement end locations and distances was examined by comparing performances of subjects with extreme scores on the space relations section of the Differential Aptitude Test. Twelve HIGHS and twelve LOWS were tested on reproduction accuracy of six distances and six end locations immediately following presentation a 30-sec rest, a 30 sec imaginal rehearsal condition, and a 30-sec imaginal distraction condition. Initial analysis of the data provided little support for the hypothesis that HIGHS would reproduce criterion locations with more accuracy than LOWS. However, when scores of HIGHS reporting use of an imaginal coding strategy (HIGHS) were compared with LOWS who did not (LOWS), several expected findings were disclosed. HIGHS reproduced end locations with significantly less error in the immediate and imaginal rehearsal conditions, and as expected, accuracy scores for HIGHS and LOWS did not differ for the distance task. The findings suggest that when subjects employ an imaginal coding strategy, visual imagery ability may be an important factor in the retention of location information, but of little functional significance in the recall of distance.     

The Temporal Placement of Interpolated Movements in Short-Term Motor Memory,

Directional response biasing as a function of the recency of an interpolated act was examined on a lever device in a within-S design. The temporal occurrence of an interpolated act was systematically varied (5, 30, or 55 sec. after the criterion act) for three retention intervals (15, 40, and 65 sec.), and the location of the interpolated movement was held constant ± 40 deg. from the criterion position. Analysis of constant errors showed significant positive directional biasing for interpolated acts greater than criterion positions. Further analyses at the 40-and 65-sec. retention intervals supported a recency interpretation. As the retention interval increased, the positive constant error was maintained for interpolated acts with the same post-biasing interval, whereas constant error became increasingly negative as the post-biasing interval increased. Examination of variable error revealed a slight increase over time and no significant differences between directional conditions. The recency findings were discussed in terms of the view that reproduction of a criterion act is dependent on the relative decay states between the interpolated and criterion traces.     

Interference with short-term visual memory produced by concurrent central processing

Double-letter memory and test stimuli were used in two experiments on a speek comparison task. Faster decision times were found when memory and test stimuli were physically identical than when they were the same in name only. This finding was true even with retention intervals as long as 12 sec and even when difficult tasks filled the retention intervals. However, the decision-time advantage of physically identical comparisons was greatest when the interval was not filled with a task likely to interfere with rehearsals. High verbal subjects had a smaller advantage for the physically identical comparisons than did low verbal subjects but were affected in the same way as low verbal subjects in terms of which conditions raised the overall correct comparison times, raised the error rates, and reduced the advantage of physically identical comparisons.     

Recognition memory retention interval and information load

Two experiments were conducted to study the effect of retention interval and information load on short-term recognition memory in retarded subjects. The task consisted of an input trial of 1–15 pictures followed by a test trial of one picture, with the subject having to press different levers to indicate whether the probe picture had been present on the input trial. Retention intervals between input trial and test trial were manipulated over a range of 1–24 sec. Both accuracy and latency of response were measured. Results showed that latency increased and accuracy decreased as a function of both retention interval and number of stimuli. In addition, number of stimuli interacted with retention interval such that accuracy was not affected by retention interval when information input was small, but decreased quite dramatically at longer retention intervals when input was larger. The results were discussed in terms of current memory search models for both normal and retarded subjects.     

Interference in memory for tonal pitch: Implications for a working-memory model

The degree of interference caused by different kinds of stimuli on memory for tonal pitch was studied. Musically trained and untrained subjects heard a sequence of two tones separated by an interval of 5 sec. The tones were either identical in pitch or differed by a semitone. Subjects had to decide whether the tones were identical or not. The interval was filled with tonal, verbal, or visual material under attended and unattended conditions. The results revealed clear group differences. Musically trained subjects' retention of the first test tone was only affected by the interposition of other tones. In contrast, the performance of musically untrained subjects was also affected by verbal and visual items. The findings are discussed in the framework of Baddeley's (1986) working-memory model.     

A test of the reward-value hypothesis

Rats retain source memory (memory for the origin of information) over a retention interval of at least 1 week, whereas their spatial working memory (radial maze locations) decays within approximately 1 day. We have argued that different forgetting functions dissociate memory systems. However, the two tasks, in our previous work, used different reward values. The source memory task used multiple pellets of a preferred food flavor (chocolate), whereas the spatial working memory task provided access to a single pellet of standard chow-flavored food at each location. Thus, according to the reward-value hypothesis, enhanced performance in the source memory task stems from enhanced encoding/memory of a preferred reward. We tested the reward-value hypothesis by using a standard 8-arm radial maze task to compare spatial working memory accuracy of rats rewarded with either multiple chocolate or chow pellets at each location using a between-subjects design. The reward-value hypothesis predicts superior accuracy for high-valued rewards. We documented equivalent spatial memory accuracy for high- and low-value rewards. Importantly, a 24-h retention interval produced equivalent spatial working memory accuracy for both flavors. These data are inconsistent with the reward-value hypothesis and suggest that reward value does not explain our earlier findings that source memory survives unusually long retention intervals.     

Counting strategies and assimilation effects in short-term retention of temporal information

The purpose of the present research was to examine proactive interference in general and assimilation effects (i.e., shifts in constant error caused by prior responses) in particular, when subjects used covert counting to aid their retention of the temporal information. Visually presented durations of 1, 4, and 8 sec. were estimated by 18 subjects under the method of reproduction. Three retention intervals (i.e., immediate, 15, and 30 sec.) and three intertrial intervals (i.e., immediate, 15, and 30 sec.) were employed. Analysis of constant error provided no indication that proactive interference was operating in the retention of temporal information as there was no increase in error across trials, no increase in error for longer retention intervals, and no interaction between trials and retention intervals. Also, there was no change in variable error as the retention intervals lengthened for any temporal duration except for the 4-sec. criterion. Finally, the rate of counting (counting units/sec.) was different across the durations to be remembered. The major conclusion of the present research was that counting greatly facilitates retention of temporal information as compared to retention without such a time-aiding strategy.     

Memory for movement: Interaction of location and distance cues and imagery ability

The present experiments examined the role of imagery ability in recall of either the terminal location or the distance of a preselected horizontal linear movement following changes in the recall starting position. Subjects were selected on the basis of their scores on a shortened version of the Betts Questionnaire upon Mental Imagery (Betts Q.M.I., Sheehan 1967). Both high and low imagers were assigned to one of four groups, HIL, LIL, HID, LID, the last letter indicating the movement cue (terminal location or distance moved) to be recalled. Recall of the appropriate movement cue from one of four new recall starting positions occurred after either a 5-sec or 30-sec unfilled retention interval. Analysis of constant error indicated all groups were unable to recall the specified movement cue independent of the other source of information, even when subjects were given explicit instructions to utilise an imagery strategy. The data corroborate our earlier findings that memory for self paced movements is based on an interaction of location and distance cues derived from the criterion movement.     

Memory Characteristics of Kinesthetic Information

Several kinesthetic cues may underlie the retention of movements: joint position receptors, muscle stretch receptors, tendon stretch receptors, cutaneous senses, duration of movements, and motor outflow all provide cues. An attempt was made to separate subsets of cues used for movement reproduction by varying the characteristics of the movements. Ss reproduced either the end Location of a movement or the Distance plus Location. The original and reproduction movements involved either the same or different muscle tensions. These manipulations failed to result in different retention characteristics. In all cases there was little loss of accuracy over a 7-sec. retention interval unless the retention interval was filled with a distracting task. These results are quite different from those of a number of other studies of movement retention, suggesting that different cues do have different retention characteristics.     

Overt is no better than covert when rehearsing visuo-spatial information in working memory

In the present study, we examined whether eye movements facilitate retention of visuo-spatial information in working memory. In two experiments, participants memorised the sequence of the spatial locations of six digits across a retention interval. In some conditions, participants were free to move their eyes during the retention interval, but in others they either were required to remain fixated or were instructed to move their eyes exclusively to a selection of the memorised locations. Memory performance was no better when participants were free to move their eyes during the memory interval than when they fixated a single location. Furthermore, the results demonstrated a primacy effect in the eye movement behaviour that corresponded with the memory performance. We conclude that overt eye movements do not provide a benefit over covert attention for rehearsing visuo-spatial information in working memory.     

The impact of object working memory on timing

Nontemporal information processing involving short-term memory requirements disturbs time estimation. Previous studies mostly used letters or digits, which are maintained in working memory by phonological loops. Since verbal and nonverbal information are processed by separate working-memory subsystems, how do nonverbal, object-based memory tasks affect time estimation? We manipulated visual object memory load using the magic cube materials. Participants were divided into three groups, who completed a reaction-time task (control task), a memory-recognition task interposed by an attempt to produce a 2500-ms time interval (active processing), and a memory-recognition task following time interval production (passive retention). The produced time increased with increasing memory-object size under both the active processing and passive retention conditions; mean produced time interval did not significantly differ between the two experimental conditions. By comparing the reaction times and error rates of a relevant task, we excluded any speed–accuracy tradeoff during timing. This result suggests that when the working-memory information to be processed includes objects requiring attention for retention, the production of time intervals is also affected by memory item maintenance.     

Imagery ability and the acquisition and retention of movements

In this study, we examined the relationship between imagery ability, as measured by the Movement Imagery Questionnaire (MIQ), and the acquisition, retention, and reacquisition of movements. Based on their MIQ scores, 10 subjects were selected for the following imagery groups: high visual/high kinesthetic (HH), high visual/low kinesthetic (HL), and low visual/low kinesthetic (LL). The subjects learned four movements to a criterion level. Before each trial, subjects kinesthetically imaged the movement about to be produced. Following each acquisition trial, subjects were provided visual feedback. The acquisition phase was followed by a 2-day retention interval, a retention test consisting of three trials on each movement (no feedback provided), and a reacquisition phase. The HH group acquired the movements in the least number of trials, the LL group required the greatest number of trials, and the HL group required an intermediate number of trials. The data for the reacquisition phase showed the same trend. There was only weak evidence for a relationship between imagery ability and the retention of the movements. These findings support the position that high imagery ability facilitates the acquisition, but probably not the short term retention, of movements.