Differential involvement of internal clock and working memory in the production and reproduction of duration: A study on older adults
Introduction
It is a well-established finding that aging is accompanied by increased inter-individual variability in cognitive functioning in various tasks (Albert et al., 1987, Hultsch et al., 2002, Morse, 1993). In this study, we took the variability characteristic of an elderly population to study the processes involved in time estimation.
In the last twenty years, research has predominantly been conducted within the framework of the scalar expectancy theory (SET) elaborated by Gibbon, Church, and Meck (1984). According to this model, temporal judgments are based on three processing stages: the clock stage consisting of time base emitting pulses which are gated to an accumulator when a switch is closed, i.e., during the time interval to be estimated; the memory stage, conceptualized as storing accumulated pulses in working memory for comparison with the content of reference memory which contains a long-term memory representation of pulses accumulated on past trials; and finally, the decision stage, in which a mechanism compares the current duration value with those stored in working or reference memory to decide on the adequate temporal response.
At the clock level, the pulse emission rate is an important factor, because the content of the accumulator provides the raw material for measuring time. Studies have shown that this rate can be manipulated by pharmacological drugs (see Meck, 1996, Rammsayer, 1999) and that it could also depend on body temperature (for a review, see Wearden & Penton-Voak, 1995). In addition, since the pioneering work of Treisman, Faulkner, Naish, and Brogan (1990), it has consistently been shown that the clock rate can be speeded up or slowed down using different rapid click-trains (Droit-Volet and Wearden, 2002, Penton-Voak et al., 1996, Wearden et al., 1999). Wearden et al. (1999, Exp. 1) applied this method to a pair-comparison procedure. In order to observe a relative slowing down, the target but not the second duration tone was preceded by clicks. In this condition, the first tone seemed to be longer than the second because more pulses were accumulated during the first tone. According to several other authors (Denner et al., 1964, Boltz, 1994), the internal time base rate is directly reflected in a person’s preferred motor tempo, which is produced by participants when asked to tap regularly at their preferred rate. Vanneste, Pouthas and Wearden’s study (2001) confirmed that the internal time base rate varies from person to person, their findings also indicating that older participants spontaneously choose a slower motor tempo than younger adults, confirming the hypothesis that the internal clock rate slows down with age. In an experiment examining the relationships between spontaneous motor tempo and remembered durations, Boltz (1994) showed that individual tapping rates and time judgments were affected when stress and relaxation were manipulated. This observation was said to show that increasing arousal led to faster internal tempos and longer estimated durations, whereas decreasing arousal led to slower internal tempos and shorter duration judgments. Several studies, using a production task in which participants had to produce durations given in conventional units of time (such as pressing a key for x seconds), have suggested that age-related differences could be due to a slowing of the internal clock (Block et al., 1998, Craik and Hay, 1999, Perbal et al., 2002). Perbal et al.’s findings (2002) indicated that longer productions in older adults were correlated to slower processing speed measures.
Working memory is also an important factor in the elaboration of a temporal judgment. Recent studies have shown that working memory and interval timing rely on the same anatomical structures (Lustig, Matell, & Meck, 2005). Studies using a mental rotation task of polygons (Fortin & Breton, 1995) and the Corkin Test (Perbal et al., 2002) as working memory tasks indicate a relationship between time estimation and working memory capacities. Moreover, studies combining temporal tasks with different memory tasks in a concurrent processing condition have shown that temporal judgment lengthened as a function of the memory-set size (Brown, 1997, Fortin and Massé, 1999, Fortin and Rousseau, 1987, Fortin and Rousseau, 1998, Fortin et al., 1993). These findings are consistent with the fact whether the material is verbal or visuospatial.
In another study, Fortin and Couture (2002) used interference between a search task in working memory and a duration reproduction task, in which participants had to reproduce a previously presented target duration, to study the relationship between short term memory and temporal processing. The memory task consisted of the presentation of several letters displayed on a screen. Two tones separated by the target duration (1.85–6.45 s) were then presented. The participant initiated the reproduction by pressing the “2” key. A probe appeared after a short time and remained on until the end of the reproduction, which was terminated by pressing the “1” or the “3” key according to the presence or absence of the probe in the memory set. The findings showed that reproductions of time were positively related to the duration of working memory processing. These findings were interpreted in terms of concurrent working memory processing demands for the two tasks: a search task and reproduction of duration. Relationships between working memory and time estimation have also been shown by Perbal et al. (2002). Regression statistical analyses revealed that shorter reproductions observed in older participants were mainly explained by working memory deficits.
The choice of method to investigate time judgments seems particularly crucial because of the putative different nature of processes involved in the various temporal tasks. Indeed, some recent studies using different groups of participants, such as elderly people or brain-damaged patients, have suggested differential involvement of the internal clock rate and memory component in duration production and reproduction tasks, the first affecting production performance and the second involved in duration reproduction (Block et al., 1998, Craik and Hay, 1999, Perbal et al., 2002, Pouthas and Perbal, 2004, Vanneste et al., 1999). Overall, these findings suggest that duration production and reproduction may be sub-served by different processes, production being more dependent on internal clock speed and reproduction more on working memory capacity.
The aim of the present study is to go further in differentiating the processes which underlie production and reproduction of durations. Participants performed these two temporal tasks, and also a rhythmic task (spontaneous motor tempo) and working memory tests. We calculated factor analysis on each measure, anticipating four distinct factors reflecting the production of a time interval, reproduction of a time interval, working memory and spontaneous motor tempo, respectively.
In order to investigate directly the differential roles of the internal time base and working memory in the duration production and reproduction tasks, we computed correlations among produced durations, reproduced durations, spontaneous motor tempo and working memory. As Moscovitch and Winocur noted (1992, p. 320): “the procedure thus permits a functional analogue of a double dissociation”: performance on Target Test 1 would be correlated with scores on test A but not test B, whereas the reverse pattern would be obtained for the Target Test 2. Recently, several neuropsychological studies have investigated the dissociation of processes, using the inter-individual variability of a group of elderly adults (Doiseau and Isingrini, 2005, Glisky et al., 1995, Nyberg et al., 1997, Souchay et al., 2004). Indeed, studies on aging largely confirmed that older people are more variable than younger ones (Hultsch et al., 2002). Moreover, age-related increased interindividual differences in cognition are observed in various tasks, such as reaction time, memory or fluid intelligence (Albert et al., 1987, Christensen et al., 1999, Hultsch et al., 2002, Morse, 1993). Hence, a factor analysis and a correlational approach might be suitable in an elderly group in order to examine a specific relationship between variables. In the present study, we therefore used the inter-individual variability of a group of elderly adults to examine the relationship between temporal tasks and spontaneous motor tempo and working memory. We assumed that production performance would be closely related to the spontaneous motor tempo rate and reproduction performance to the working memory scores.
Section snippets
Participants
A total of 56 right-handed elderly adults, 40 females and 16 males (age range 60–96 years, mean age: 74.45, SD: 8.82), participated in this experiment. They had at least 8 years of schooling (M: 11.21, SD: 2.58) and their cultural level was assessed by the vocabulary score of the Mill Hill test (M: 26.59, SD: 4.45) (Raven, 1982). All participants reported themselves to be in good physical and mental health and free from medication known to affect the central nervous system. The older adults
Results
Descriptive statistics (i.e., means and standard deviations) for all measures (production, reproduction, spontaneous tempo and working memory) are reported in Table 1. First, in order to check whether or not there was an improvement in temporal performance for the later trials, ANOVAs (2 durations × 3 trials) were carried out on mean estimations for both production and reproduction tasks. No effect of the trial factor was found neither for the production task nor for the reproduction task (F(2,
Discussion
The aim of the present study was to differentiate further the processes which underlie production and reproduction of durations. Findings of the factor analysis conducted on temporal performance, rhythmic performance and working memory measures revealed four distinct factors separating duration production and reproduction, spontaneous motor tempo measures and working memory performances. Correlational analysis indicated significant correlations between pairs of factors: duration production and
References (46)
- et al.
Mini-Mental State: A practical method for grading the cognitive state of patients for the clinician
Journal of Psychiatric Research
(1975) Time reproductions by H.M
Acta Psychologica
(1973)- et al.
Speeding up and (…relatively…) slowing down an internal clock in humans
Behavioral Process
(1999) - et al.
Nonlinear changes in cognition with age and their neuropsychologic correlates
Canadian Journal of Psychology
(1987) - et al.
Human aging and duration judgments: A meta-analytic review
Psychology and Aging
(1998) Changes in internal tempo and effects on the learning and remembering of event durations
Journal of Experimental Psychology: Learning, Memory and Cognition
(1994)Attentional resources in timing: Interference effects in concurrent temporal and nontemporal working memory tasks
Perception and Psychophysics
(1997)- et al.
An analysis of diversity in the cognitive performance of elderly community dwellers: Individual differences in change scores as a function of age
Psychology and Aging
(1999) Properties of the internal clock
- et al.
Aging and judgments of duration: Effects of task complexity and method of estimation
Perception and Psychophysics
(1999)