时间分层加工研究述评

为了探讨数秒内不同层级时间加工的特性,研究者分别从“时间信息加工”和“信息加工的计时特性”角度开展了一系列研究。Münsterberg (1889)、Michon(1985) 、Lewis 和 Miall(2003) 及Vierodt(1868)从前一角度,分别指出1/3秒、1/2秒、1秒及3秒可能是数秒以内时距加工机制的分界点,分界点以下与以上的加工机制存在差异。P?ppel(1997, 2009)则从后一角度指出限制信息加工过程的两类时间窗,一类时间窗是以20-60毫秒振荡周期运行的高频系统,属于初级整合单元;另一类时间窗主要是处理2-3秒以内事件系列的低频系统,属于高级整合单元。前一类时间窗可以为信息加工整合基本的心理事件,后一类时间窗则是把2-3秒内的心理事件整合为基本知觉单元。基于以往研究的剖析,我们认为1/3秒、1/2秒及1秒等分界点的真伪性尚需进一步验证,并进一步假设40毫秒以内时间不能觉察为时距;40毫秒至3秒之间,随着长度增加,自动化加工减弱,控制性加工增强;3秒以上主要为控制性加工,涉及记忆过程。     

The lifespan of time intervals in reference memory

To further explore how memory influences time judgments, we conducted two experiments on the lifespan of temporal representations in memory. Penney et al (2000, Journal of Experimental Psychology Human Perception and Performance 26 1770-1787) reported that the perceived duration of auditorily and visually marked intervals differs only when both marker-type intervals are compared directly. This finding can be explained by a 'memory-mixing' process, whereby the memory trace of previous intervals influences the perception of upcoming ones, which are then added to the memory content. In the experiments discussed here, we manipulated the mixing mode of auditory/visual signal presentations. In experiment 1, signals from the same modality were either grouped by blocks or randomised within blocks. The results showed that the auditory/visual difference decreased but remained present when modalities were grouped by blocks. In experiment 2, we used a line-segmentation task. The results showed that, after a training block was performed in one modality, the perceived duration of signals from the other modality was distorted for at least 30 trials and that the magnitude of the difference decreased as the block went on. The results of both experiments highlight the influence of memory on time judgments, providing empirical support to, and quantitative portrayal of, the memory-mixing process.     

Temporal reproduction: further evidence for two processes

Some authors have suggested separate mechanisms for the processing of temporal intervals above versus below 2-3s. Given that the evidence is mixed, the present experiment was carried out as a critical test of the separate-mechanism hypothesis. Subjects reproduced five standard durations of 1-5s presented in the auditory and visual modalities. The Corsi-block test was used to assess effects of working-memory span on different interval lengths. Greater working-memory span was associated with longer reproductions of intervals of 3-5s. A factor analysis run on mean reproduced intervals revealed one modality-unspecific factor for durations of 1-2s and two modality-specific factors for longer intervals. These results are interpreted as further indications that two different processes underlie temporal reproductions of shorter and longer intervals.     

Testing the representation of time in reference memory in the bisection and the generalization task: The utility of a developmental approach

This study examined the effect of the variability of representation of durations in reference memory on temporal discrimination performance in children aged 5 and 8 years as well as in adults using a bisection (Experiment 1) and a generalization task (Experiment 2). In each task, the participants were familiarized before the blocks of tested trials with either the same referent duration values (fixed condition) or a distribution of referent duration values, with a mean equal to the referent durations used in the fixed condition and a .20 coefficient of variation (variable condition). The results showed that the sensitivity to duration was lower in the variable than in the fixed condition in the children and, to a lesser extent, in the adults. The modelling of the data indicated that this effect was due to the increase in the variability of the representation of durations in reference memory, but also to changes in the decisional processes.     

Testing the representation of time in reference memory in the bisection and the generalization task: the utility of a developmental approach

This study examined the effect of the variability of representation of durations in reference memory on temporal discrimination performance in children aged 5 and 8 years as well as in adults using a bisection (Experiment 1) and a generalization task (Experiment 2). In each task, the participants were familiarized before the blocks of tested trials with either the same referent duration values (fixed condition) or a distribution of referent duration values, with a mean equal to the referent durations used in the fixed condition and a .20 coefficient of variation (variable condition). The results showed that the sensitivity to duration was lower in the variable than in the fixed condition in the children and, to a lesser extent, in the adults. The modelling of the data indicated that this effect was due to the increase in the variability of the representation of durations in reference memory, but also to changes in the decisional processes.     

Short-term Memory in Time Interval Production

This article summarizes results from several studies on the effect of nontemporal processing on concurrent time estimation. These studies use an experimental paradigm that was developed to interpolate various nontemporal tasks in time interval production. Produced intervals were specifically lengthened by increasing the amount of short-term memory processing performed in a concurrent search task. The interference was shown with many short-term memory tasks, using various durations and methods in the temporal task. The effect on temporal production seems to be positively related to the level of difficulty of the short-term memory task. Finally, processing item or order information affects simultaneous time production, whereas its maintenance has no effect. An interference analysis of these phenomena suggests that interruption in time estimation is caused by both temporal and nontemporal tasks simultaneously requiring processing in short-term memory.     

Elaborative rehearsal of nontemporal information interferes with temporal processing of durations in the range of seconds but not milliseconds

The distinct timing hypothesis suggests a sensory mechanism for processing of durations in the range of milliseconds and a cognitively controlled mechanism for processing of longer durations. To test this hypothesis, we employed a dual-task approach to investigate the effects of maintenance and elaborative rehearsal on temporal processing of brief and long durations. Unlike mere maintenance rehearsal, elaborative rehearsal as a secondary task involved transfer of information from working to long-term memory and elaboration of information to enhance storage in long-term memory. Duration discrimination of brief intervals was not affected by a secondary cognitive task that required either maintenance or elaborative rehearsal. Concurrent elaborative rehearsal, however, impaired discrimination of longer durations as compared to maintenance rehearsal and a control condition with no secondary task. These findings endorse the distinct timing hypothesis and are in line with the notion that executive functions, such as continuous memory updating and active transfer of information into long-term memory interfere with temporal processing of durations in the second, but not in the millisecond range.     

Neuropharmacological Evidence for Different Timing Mechanisms in Humans

Temporal processing of intervals in the range of seconds or more is cognitively mediated, whereas processing of brief durations below 500 msec appears to be based on brain mechanisms outside cognitive control. To elucidate the critical role of various neurotransmitters in timing processes in humans, the effects of 3 mg of haloperidol, a dopamine receptor antagonist, 11 mg of the benzodiazepine midazolam, and 1 mg of scopolamine, a cholinergic receptor antagonist, were compared in a placebo-controlled double-blind experiment. In addition, changes in cortical arousal, semantic memory, and cognitive and motor skill acquisition were assessed. Temporal processing of long durations was significantly impaired by haloperiodol and midazolam, whereas processing of extremely brief intervals was only affected by haloperidol. The overall pattern of results supports the notion that temporal processing of longer intervals is mediated by working-memory functions and, therefore, any pharmacological treatment, irrespective of the neurotransmitter system involved, that produces a deterioration of working memory, may interfere with temporal processing of longer intervals. Temporal processing of intervals in the range of milliseconds appears to depend on the effective level of dopaminergic activity in the basal ganglia.     

When do auditory/visual differences in duration judgements occur?

Four experiments examined judgements of the duration of auditory and visual stimuli. Two used a bisection method, and two used verbal estimation. Auditory/visual differences were found when durations of auditory and visual stimuli were explicitly compared and when durations from both modalities were mixed in partition bisection. Differences in verbal estimation were also found both when people received a single modality and when they received both. In all cases, the auditory stimuli appeared longer than the visual stimuli, and the effect was greater at longer stimulus durations, consistent with a “pacemaker speed” interpretation of the effect. Results suggested that Penney, Gibbon, and Meck's (2000) “memory mixing” account of auditory/visual differences in duration judgements, while correct in some circumstances, was incomplete, and that in some cases people were basing their judgements on some preexisting temporal standard.     

Lesions in the anterior thalamic nuclei of rats do not disrupt acquisition of stimulus sequence learning

Four experiments examined the effects of encoding multiple standards in a temporal generalization task in the visual and auditory modalities both singly and cross-modally, using stimulus durations ranging, across different experiments, from 100 to 1,400 ms. Previous work has shown that encoding and storing multiple auditory standards of different durations resulted in systematic interference with the memory of the standard, characterized by a shift in the location of peak responding, and this result, from Ogden, Wearden, and Jones (2008), was replicated in the present Experiment 1. Experiment 2 employed the basic procedure of Ogden et al. using visual stimuli and found that encoding multiple visual standards did not lead to performance deterioration or any evidence of systematic interference between the standards. Experiments 3 and 4 examined potential cross-modal interference. When two standards of different modalities and durations were encoded and stored together there was also no evidence of interference between the two. Taken together, these results, and those of Ogden et al., suggest that, in humans, visual temporal reference memory may be more permanent than auditory reference memory and that auditory temporal information and visual temporal information do not mutually interfere in reference memory.     

No evidence for qualitative differences in the processing of short and long temporal intervals

Several lines of research suggest that two distinct timing mechanisms are involved in temporal information processing: a sensory mechanism for processing of durations in the range of milliseconds and a cognitively controlled mechanism for processing of longer durations. The present study employed a dual-task approach and a sensory interference paradigm to further elucidate the distinct timing hypothesis. Experiment 1 used mental arithmetic as a nontemporal secondary task, Experiment 2 a memory search task, and Experiment 3 a visuospatial memory task. In Experiment 4, a loudness manipulation was applied. Mental arithmetic and loudness manipulation affected temporal discrimination of both brief and long intervals, whereas the two remaining tasks did not influence timing performance. Observed differences in interference patterns may be explained by some tasks being more difficult than others. The overall pattern of results argues against two qualitatively distinct timing mechanisms, but is consistent with attention-based cognitive models of human timing.     

Modality effects in memory for basic stimulus attributes: a temporal and nontemporal comparison

Previous research suggests that there are significant differences in the operation of reference memory for stimuli of different modalities, with visual temporal entries appearing to be more durable than auditory entries (Ogden, Wearden, & Jones, 2008 , 2010). Ogden et al. ( 2008 , 2010 ) demonstrated that when participants were required to store multiple auditory temporal standards over a period of delay there was significant systematic interference to the representation of the standard characterized by shifts in the location of peak responding. No such performance deterioration was observed when multiple visually presented durations were encoded and maintained. The current article explored whether this apparent modality-based difference in reference memory operation is unique to temporal stimuli or whether similar characteristics are also apparent when nontemporal stimuli are encoded and maintained. The modified temporal generalization method developed in Ogden et al. (2008) was employed; however, standards and comparisons varied by pitch (auditory) and physical line length (visual) rather than duration. Pitch and line length generalization results indicated that increasing memory load led to more variable responding and reduced recognition of the standard; however, there was no systematic shift in the location of peak responding. Comparison of the results of this study with those of Ogden et al. (2008, 2010) suggests that although performance deterioration as a consequence of increases in memory load is common to auditory temporal and nontemporal stimuli and visual nontemporal stimuli, systematic interference is unique to auditory temporal processing.     

Short-term memory and time estimation: beyond the 2-second "critical" value.

We showed previously that when time intervals around two seconds (s) are reproduced concurrently with a memory task, intervals are positively related to duration of memory processing. However, some data in research on timing as well as in memory research suggest that 2 s might be a critical duration beyond which different mechanisms or structures would support performance. This implies that the interference observed between memory processing and 2-s productions could be specific to these durations, and would not be obtained with longer durations. In this experiment, intervals ranging from 1.85 to 6.45 s were reproduced by participants, who were searching simultaneously for a memory probe. At all durations, reproductions were positively related to memory set size. These findings have implications with regards to previous research indicating a discontinuity around 2-3 s in time perception. They suggest in particular that the role of memory is similar in reproduction of durations around 2 s and of longer durations.     

Temporal interval production and processing in working memory

Short-term memory or working memory has been proposed as a cognitive structure contributing to time estimation. Thus, in a previous experiment, retrieving a stored item during a temporalinterval production lengthened the interval in proportion to the number of items in the memory set. In the present study, this issue was analyzed further by testing whether the proportional lengthening is induced by the load itself (i.e., the number of items) or by comparing the probe with memorized items. In a first experiment, a memory set was maintained during a temporal production, and the comparison of the probe with memorized items was postponed until the end of time production. Varying the number of items in the memory set had no effect on temporal intervals produced during its retention, suggesting that mental comparison was the source of the lengthening of time intervals. In succeeding experiments, tasks requiring processing in working memory but involving no memory load were combined with temporal production. In Experiment 2, increasing the number of syllables in a rhyme-judgment task proportionally lengthened temporal intervals that were produced simultaneously. In Experiment 3, increasing the amount of mental rotation in a task involving visuospatial processing also lengthened simultaneous temporal production. This interference between processing in working memory and time estimation suggests that working memory, defined as a work space for active processing of current information, contributes to time estimation.     

Auditory and visual differences in time perception? An investigation from a developmental perspective with neuropsychological tests

Adults and children (5- and 8-year-olds) performed a temporal bisection task with either auditory or visual signals and either a short (0.5-1.0s) or long (4.0-8.0s) duration range. Their working memory and attentional capacities were assessed by a series of neuropsychological tests administered in both the auditory and visual modalities. Results showed an age-related improvement in the ability to discriminate time regardless of the sensory modality and duration. However, this improvement was seen to occur more quickly for auditory signals than for visual signals and for short durations rather than for long durations. The younger children exhibited the poorest ability to discriminate time for long durations presented in the visual modality. Statistical analyses of the neuropsychological scores revealed that an increase in working memory and attentional capacities in the visuospatial modality was the best predictor of age-related changes in temporal bisection performance for both visual and auditory stimuli. In addition, the poorer time sensitivity for visual stimuli than for auditory stimuli, especially in the younger children, was explained by the fact that the temporal processing of visual stimuli requires more executive attention than that of auditory stimuli.     

Duration discrimination of empty and filled intervals marked by auditory and visual signals

Experiments 1 and 2 compared, with a single-stimulus procedure, the discrimination of filled and empty intervals in both auditory and visual modalities. In Experiment 1, in which intervals were about 250 msec, the discrimination was superior with empty intervals in both modalities. In Experiment 2, with intervals lasting about 50 msec, empty intervals showed superior performance with visual signals only. In Experiment 3, for the auditory modality at 250 msec, the discrimination was easier with empty intervals than with filled intervals with both the forced-choice (FC) and the single stimulus (SS) modes of presentation, and the discrimination was easier with the FC than with the SS method. Experiment 4, however, showed that at 50 and 250 msec, with a FC-adaptive procedure, there were no differences between filled and empty intervals in the auditory mode; the differences observed with the visual mode in Experiments 1 and 2 remained significant. Finally, Experiment 5 compared differential thresholds for four marker-type conditions, filled and empty intervals in the auditory and visual modes, for durations ranging from .125 to 4 sec. The results showed (1) that the differential threshold differences among marker types are important for short durations but decrease with longer durations, and (2) that a generalized Weber’s law generally holds for these conditions. The results as a whole are discussed in terms of timing mechanisms.     

Interference between auditory and visual duration judgements suggests a common code for time

Auditory stimuli usually have longer subjective durations than visual ones for the same real duration, although performance on many timing tasks is similar in form with different modalities. One suggestion is that auditory and visual stimuli are initially timed by different mechanisms, but later converted into some common duration code which is amodal. The present study investigated this using a temporal generalization interference paradigm. In test blocks, people decided whether comparison durations were or were not a 400-ms standard on average. Test blocks alternated with interference blocks where durations were systematically shorter or longer than in test blocks, and interference was found, in the direction of the durations in the interference blocks, even when the interfering blocks used stimuli in a different modality from the test block. This provides what may be the first direct experimental evidence for a “common code” for durations initially presented in different modalities at some level of the human timing system.     

Spatial frequency filtering and the direct control of fixation durations during scene viewing

The present study employed a saccade-contingent change paradigm to investigate the effect of spatial frequency filtering on fixation durations during scene viewing. Subjects viewed grayscale scenes while encoding them for a later memory test. During randomly chosen saccades, the scene was replaced with an alternate version that remained throughout the critical fixation that followed. In Experiment 1, during the critical fixation, the scene could be changed to high-pass and low-pass spatial frequency filtered versions. Under both conditions, fixation durations increased, and the low-pass condition produced a greater effect than the high-pass condition. In subsequent experiments, we manipulated the familiarity of scene information during the critical fixation by flipping the filtered scenes upside down or horizontally. Under these conditions, we observed lengthening of fixation durations but no difference between the high-pass and low-pass conditions, suggesting that the filtering effect is related to the mismatch between information extracted within the critical fixation and the ongoing scene representation in memory. We also conducted control experiments that tested the effect of changes to scene orientation (Experiment 2a) and the addition of color to a grayscale scene (Experiment 2b). Fixation distribution analysis suggested two effects on the distribution fixation durations: a fast-acting effect that was sensitive to all transsaccadic changes tested and a later effect in the tail of the distribution that was likely tied to the processing of scene information. These findings are discussed in the context of theories of oculomotor control during scene viewing.     

The role of the human medial temporal lobe in object recognition and object discrimination.

This paper reviews evidence from neuropsychological patient studies relevant to two questions concerning the functions of the medial temporal lobe in humans. The first is whether the hippocampus and the adjacent perirhinal cortex make different contributions to memory. Data are discussed from two patients with adult-onset bilateral hippocampal damage who show a sparing of item recognition relative to recall and certain types of associative recognition. It is argued that these data are consistent with Aggleton and Brown's (1999) proposal that familiarity-based recognition memory is not dependent on the hippocampus but is mediated by the perirhinal cortex and dorso-medial thalamic nucleus. The second question is whether the recognition memory deficit observed in medial temporal lobe amnesia can be explained by a deficit in perceptual processing and representation of objects rather than a deficit in memory per se. The finding that amnesics were impaired at recognizing, after short delays, patterns that they could successfully discriminate suggests that their memory impairment did not result from an object-processing deficit. The possibility remains, however, that the human perirhinal cortex plays a role in object processing, as well as in recognition memory, and data are presented that support this possibility.     

Cued shifts of attention and memory encoding in partial report: A dual-task approach

This study explores how cued shifts of visual attention and rapid encoding of visual information relate to limited-capacity processing mechanisms. Three experiments were conducted placing a partial-report task within a dual-task paradigm. Experiments 1 and 2 involved a simple speeded visual discrimination (Task 1) and then an unspeeded partial-report task (Task 2). Generally, Task 2 accuracy declined as the temporal overlap between the two tasks increased. In addition, in Experiment 1, varying the number of items in the partial-report display had an effect on performance regardless of overlap. In contrast, in Experiment 2, varying the type of probe had an effect only at long task overlap. The generality of the interference effect was tested in Experiment 3 using an auditory discrimination as Task 1. Again, Task 2 accuracy declined as the temporal overlap between the two tasks increased. In all cases, the observed interference had the properties of a processing bottleneck. It is argued that encoding information into memory and response selection for the first task both require general-purpose processing. The results are discussed in terms of the functional relationship between attention and memory.