Adam, J.J., Jakob, R., Bovend’Eerdt, T.J.H., & Van Gerven, P.W.M. (2012). Spared within-hands but impaired between-hands response preparation in aging. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 67(3), 317–324, doi:10.1093/geronb/gbr105. Advance Access published on September 14, 2011
Spared Within-Hands but Impaired Between-Hands Response Preparation in Aging Jos J. Adam,1 Rolf Jakob,1 Thamar J. H. Bovend’Eerdt,1 and Pascal W. M. Van Gerven2
2Faculty
1Faculty of Health, Medicine, and Life Sciences, of Psychology and Neuroscience, Maastricht University, The Netherlands.
Objectives. Older people can use advance information to prepare a subset of finger responses. It is debated, however, whether aging affects the preparation of finger responses on two hands (between-hands preparation) more strongly than the preparation of finger responses on one hand (within-hands preparation). The present study examined the role of temporal uncertainty in this issue.
Results. Reaction time and error rates revealed age equivalence for within-hands preparation but an age-related difference for between-hands preparation, regardless of how the preparation intervals were presented. Discussion. These findings demonstrate a robust, structural difference in the maximal preparation benefit that older adults can achieve when preparing two fingers on two hands but not on one hand. These outcomes are discussed in terms of several theories of cognitive aging. Key Words: Aging—Context Processing—Reaction Time—Response Preparation—Temporal Uncertainty.
Background It is well known that older people can use advance information delivered by cues to prepare their responses, even though they may need more time to do this than younger adults (for a review, see Proctor, Vu, & Pick, 2005). It remains unclear, however, whether older people can attain the same “maximal preparation benefit” (defined as the largest reduction in reaction time (RT) in a cued, prepared condition compared with an uncued, unprepared condition) as younger adults do, even when provided with sufficiently long preparation time. The available evidence is scant and mixed. One study reported evidence for an age-related difference in the maximal preparation benefit that can be attained (Proctor, Vu, & Pick, 2006, Experiment 2), whereas another study reported no such difference (Moresi et al., 2009). It is important to note that these discrepant findings concern the maximal preparation benefit that can be achieved when preparing two fingers on two hands, not the maximal preparation benefit when preparing two fingers on one hand, as within-hands preparation often shows no age effect (Adam et al., 1998; Moresi et al., 2009; Proctor et al., 2006; for an exception, see Sterr & Dean, 2008). One salient factor that distinguishes the conflicting studies on between-hands preparation is temporal uncertainty. The study that reported evidence for an age-related difference (Proctor et al., 2006) presented the different preparation intervals in isolation, that is, in separate blocks of trials (i.e., a fixed design). The study that reported evidence against an age-related difference (Moresi et al., 2009) presented
the preparation intervals randomly intermixed (i.e., mixed design). In the present study, we examined the role of fixed versus mixed preparation intervals in the maximal preparation benefit that older people can achieve when preparing two fingers on different hands as opposed to preparing two fingers on the same hand. Before explicating this aim and its theoretical significance, we first describe our experimental paradigm (finger-cuing task) and a recent account of fingercuing effects (Grouping Model).
The Finger-Cuing Task Miller (1982) devised the finger-cuing task to study the preparation of discrete finger responses. In this task, three horizontal rows of symbols (representing warning, cue, and target signal) appear consecutively and underneath each other in the center of a computer monitor (see Figure 1). At the beginning of a trial, the warning signal (a row of four plus signs) appears for 500 ms. After that, the cue signal (containing two plus signs) appears in a separate row, indicating a subset of two possible finger responses. Next, after a certain preparation interval, a single plus sign appears, signaling the required response. The response is a spatially compatible key press, executed by the index or middle finger of the left or right hand. The functional significance of the cue signal is to specify a subset of two (out of four) possible stimulus–response locations, thus converting the basic 4-choice task into a 2-choice task. Four cue types can be distinguished (see Figure 1). In
© The Author 2011. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail:
[email protected]. Received February 3, 2011; Accepted August 23, 2011 Decision Editor: Robert West, PhD
317
Downloaded from http://psychsocgerontology.oxfordjournals.org/ by guest on January 7, 2016
Methods. We asked a group of young and older participants to perform a finger-cuing task with four preparation intervals (2, 3, 4, and 5 s), presented either separately in distinct blocks of trials (fixed design: no temporal uncertainty) or randomly intermixed across trials (mixed design: temporal uncertainty).
318
ADAM ET AL.
Warning Cue Stimulus
Uncued
Hand-Cued
Finger-Cued
Neither-Cued
+ + + + + + + + + + + +
+ + + + + + + + +
+ + + + + + + + + + + +
+ + + + + + + + + + + +
+ + + + + + + + + + + +
+ + + + + + + + + + + +
+ + + + + + + + + + + +
Prepared Responses
Figure 1. The four cue types in the finger-cuing task developed by Miller (1982).
Grouping Model In accounting for the hand superiority effect at short preparation intervals, Adam, Hommel, and Umiltà (2003, 2005) recently proposed the Grouping Model, which is an extension of the salient features coding principle advanced
by Proctor and Reeve (1986; Reeve & Proctor, 1990; for related notions, see Klapp & Jagacinski, 2011). The basic idea of the Grouping Model is that the individual elements of multielement visual displays and multielement response arrays are not processed independently but are pre-attentively organized or “grouped” according to low-level grouping factors that depend on stimulus-driven factors (e.g., Gestalt principles) and on response-related factors (e.g., inter response tendencies). That is, each stimulus set and each response set has a default organization established preattentively by the bottom–up computation of perceptual and motor units or subgroups; this process is fast and automatic. With additional, top–down processing, however, alternative organizations can be attained; this process is slow and effortful. Thus, the pattern of cuing effects that emerge in the finger-cuing task critically depends on the nature of these default groupings and on the time available to reorganize these representations, if necessary. According to the Grouping Model, the processing advantage of hand cues simply reflects the natural, strong grouping of the two leftmost and two rightmost stimulus– response elements that leads to a fast, automatic activation of fingers on the same hand. The coactivation of adjacent areas of the motor cortex corresponding to fingers from the same hand could be at the basis of this grouping. In contrast, bilateral finger and neither cues contain stimulus– response elements on both sides of the perceptual–motor work space that are not easily coded as belonging to a group. This is because these elements intrinsically belong to different groups (i.e., the left–right or hand groups). Hence, top–down controlled processing is needed to break or overrule the low-level, bottom–up formation of left– right perceptual–motor subgroups. Thus, when grouping is weak, ambiguous, or complex—as is the case with the finger and neither cues—it must be supported by a slow, effortful top–down process. In this view, within-hands preparation proceeds fast and automatically because it exploits the neuroanatomical hand/hemisphere distinction. Between-hands preparation, in contrast, requires nonautomatic, effortful processes to bind and prepare finger responses across hands/hemispheres.
Downloaded from http://psychsocgerontology.oxfordjournals.org/ by guest on January 7, 2016
the hand-cued condition, the cue specifies two fingers on the same hand (e.g., the left middle and left index fingers). In the finger-cued condition, the cue specifies the same finger on two hands (e.g., the left index and right index fingers). In the neither-cued condition, the cue specifies different fingers on two hands (e.g., the left index and right middle fingers). These three preparation conditions are called the “cued” or “informative” conditions. Also, an “uncued” or “uninformative” condition is included. In this control condition, no advance information about the upcoming response is provided (the four plus signs are repeated), so that selective preparation of any combination of two finger responses is not possible. In other words, the uncued condition leaves the basic 4-choice task unaltered and, thus, is a control condition against which the effects of the cued conditions can be evaluated. Because 2-choice tasks yield substantially shorter RTs than 4-choice tasks (Hick, 1952; Hyman, 1953), cue effectiveness is inferred from a significant RT advantage or benefit for the 2-choice cued conditions (i.e., hand, finger, and neither cued) relative to the 4-choice uncued condition. The robust finding from the finger-cuing paradigm is a pattern of differential cuing benefits: RTs are shortest for the hand-cued condition, longest for the neither-cued condition, and intermediate for the finger-cued condition, reflecting an ordering in terms of preparation difficulty (Reeve & Proctor, 1984, 1990). Importantly, this pattern of differential cuing benefits is only apparent at short preparation intervals (i.e., intervals less than about 2 s). Longer preparation intervals usually show a pattern of equivalent cuing benefits. Thus, responses can be selected and prepared more quickly when they are grouped together on one hand than when they are distributed over two hands, with no differences between the three cue types given sufficiently long preparation time (for a review, see Reeve & Proctor, 1990).
RESPONSE PREPARATION IN AGING
Purpose of Present Study The primary aim of the present study was to provide an empirical demonstration of the role of temporal uncertainty in shaping the maximal preparation benefit that older people
can achieve when preparing two fingers on the same hand or different hands. We used a sensitive within-participants design, requiring a group of younger and older participants to perform the finger-cuing task both with a fixed and mixed preparation interval design. Of particular interest was the efficiency of within- compared with between-hands preparation as a function of temporal uncertainty. Based on previous findings, we tested the hypothesis that older adults would exhibit a difference in between-hands preparation with fixed preparation intervals, but not with mixed preparation intervals. Furthermore, we used four long preparation intervals (2, 3, 4, and 5 s) to allow ample time for preparation and to be confident that a possible age-related difference in between-hands preparation could not be attributed to insufficient preparation time. In fact, the 2- to 5-s range substantially outlasts the critical value of 2 s, which, for younger adults, has been shown to abolish RT differences between hand-, finger-, and neither-cued conditions (Adam, Hommel, & Umiltà, 2003; Reeve & Proctor, 1990). Hence, for younger participants, we predicted a pattern of equivalent precuing benefits, regardless of the length of the preparation interval. Method
Participants Fifty healthy individuals with no history of ophthalmological or neurological problems were tested from an opportunistic sample. Participants were split in two groups: 23 younger adults (age range: 18–25 years; M = 22.2 years, 11 females) and 27 older adults (age range: 59–79 years, M = 65.7 years, 13 females). The experiment was approved by the Ethical Committee of the Faculty of Psychology and Neuroscience at Maastricht University, and all participants gave written informed consent in accordance with the Declaration of Helsinki. The two age groups were not reliably different in terms of years of education (M values = 16.9 and 15.4 for younger and older adults, respectively), t(48) = 1.69, P > .05. Stimuli and Apparatus Participants were seated in front of a 15-inch computer monitor. Viewing distance was approximately 50 cm. Responses were made by pressing one of four adjacent keys in the middle of the bottom row of a standard keyboard (V, B, N, M). Stimuli were plus (+) signs of the standard ASCII character set. Each plus sign was approximately 6 mm wide and 3 mm high (0.69 and 0.34 degrees of visual angle, respectively). Stimuli were presented in the middle of the display and consisted of a warning signal, a cue signal, and a target stimulus (see Figure 1). The warning signal consisted of four plus signs separated by 15 mm. After a delay of 500 ms, the cue signal appeared exactly below the warning signal. The cue consisted of plus signs in either all four
Downloaded from http://psychsocgerontology.oxfordjournals.org/ by guest on January 7, 2016
Temporal Uncertainty and Response Preparation in Aging Temporal preparation refers to the synchronization of a response in time. It is a critical aspect of preparatory behavior that can improve RT (e.g., Bertelson, 1967), but less so in older people (e.g., Welford, 1977). Three types of deficient temporal preparation in older age have been reported: (a) an inability to develop an optimal prepared state rapidly (e.g., Proctor et al., 2006); (b) an inability to maintain preparation over a longer period of time (e.g., Gottsdanker, 1982); and (c) an inability to prepare for uncertain events (e.g., Lahtela, Niemi, & Kuusela, 1985). An important procedural factor in the study of temporal preparation is the use of a fixed or mixed design (Bherer & Belleville, 2004; Los, 1996). In a fixed design, the same preparation interval is used consistently within a block of trials, whereas in a mixed design, the preparation interval varies unpredictably from trial to trial. Moresi colleagues (2009) have suggested that this factor may underlie the divergent findings in the literature concerning the maximal preparation benefit that can be attained when preparing two fingers on different hands. The study that reported evidence for an age-related difference in between-hands preparation used a fixed design (Proctor et al., 2006), whereas the study that reported evidence against an age-related difference used a mixed design (Moresi et al., 2009). With a fixed design, participants know in advance how much time there is to select and prepare the cued finger responses before the target appears. With a mixed design, in contrast, participants do not know how much time is available for response preparation. Hence, with a mixed design, there is more temporal uncertainty, which may make preparation less efficient, especially at shorter preparation intervals. Indeed, with a mixed design, RT generally decreases as a function of preparation interval, whereas with a fixed design, RT generally increases. This pattern is robust, as demonstrated by many studies (for a review, see Niemi & Näätänen, 1981), including those by Proctor colleagues (2006) and Moresi colleagues (2009). The first phenomenon has been interpreted as indicating that participants increase their level of preparation with increasing preparation interval because as the preparation interval lengthens the probability that the target will appear increases also (e.g., Stilitz, 1972). The second phenomenon is commonly interpreted as reflecting a difficulty in maintaining a general readiness, or set, to respond (e.g., Gottsdanker, 1975). Assuming that a fixed design introduces less temporal uncertainty than a mixed design, this factor might modulate the superiority of within-hands preparation over between-hands preparation in older age.
319
ADAM ET AL.
Analysis RTs less than 150 ms or in excess of 1,500 ms (young: 0.10%; old: 0.61% of all trials) were considered outliers and excluded from data analysis. Mean (correct) RT and percentage of errors were calculated for each participant as a function of presentation mode, cue type, and preparation interval. A mixed-model analysis of variance (ANOVA) was performed on mean RT and on the arcsine-transformed error rates (to attain normality of distributions and homogeneity of variance), with age (young/old) as between-groups variable, and preparation interval (2, 3, 4, and 5 s), presentation mode (mixed and fixed), and cue type (uncued, hand cued, finger cued, and neither cued) as within-groups variables. We also performed an ANOVA on log-transformed RTs to correct for generalized age-related slowing effects (Faust, Balota, Spieler, & Ferraro, 1999), but the results of this analysis did not deviate from those of the untransformed RT analysis and therefore are not reported. Results
Reaction Time Mean RT as a function of age, cue type, and preparation interval is presented in Figure 2a for the fixed presentation
FIXED PREPARATION INTERVALS YOUNG
800
OLD
700 600
Uncued Hand Finger Neither
500 400 300
2
3
4
5
2
3
4
5
Preparation Interval (s)
B
MIXED PREPARATION INTERVALS YOUNG
800
OLD
700 600
Uncued Hand Finger Neither
500 400 300 2
3
4
5
2
3
4
5
Preparation Interval (s) Figure 2. Mean reaction time (milliseconds) of the older and younger participants as a function of preparation interval and cue type for (a) fixed and (b) mixed preparation intervals.
mode and in Figure 2b for the mixed presentation mode. The results of the ANOVA revealed that all main effects were highly significant, indicating longer RTs for older (M = 630 ms) than for younger participants (M = 452 ms), F(1,48) = 50.15, P < .001, longer RTs for mixed (M = 551 ms) than for fixed preparation intervals (M = 531 ms), F(1,48) = 10.83, P < .001, longer RTs for the shortest preparation interval of 2 s (M = 553 ms) than for the longer ones (M values = 534, 540, and 536 ms, respectively), F(3,144) = 8.97, P < .001, and shorter RTs for all three informative cue conditions (M hand cued = 516 ms, M finger cued = 529 ms, M neither cued = 532 ms) relative to the uncued control condition (M = 586 ms), F(3,144) = 62.65, P < .001. These main effects, however, were qualified by four significant 2-way interactions. The interactions involving three or more factors were not significant (all P values > .1). The cue type × preparation interval interaction, F(9,432) = 4.09, P < .001, indicated greater preparation benefits for all three informative cue conditions (hand, finger, and neither cued) with the three longest preparation intervals compared with the shortest. The cue type × presentation mode
Downloaded from http://psychsocgerontology.oxfordjournals.org/ by guest on January 7, 2016
Design and Procedure All participants performed the finger-cuing task with the two presentation modes (fixed vs. mixed preparation intervals) administered in one single session. The order of the presentation mode of the preparation intervals was counterbalanced. In the fixed presentation condition, participants received four series of 80 trials, one for each of the four preparation intervals (2, 3, 4, and 5 s). The order of preparation intervals was random. Within a series of 80 trials, there were 20 trials for each of the four cue conditions, presented in random order. In the mixed presentation condition, participants also received four blocks of each 80 experimental trials. However, in this condition, the order of the four preparation intervals (as well as the order of the four cue conditions) varied randomly across trials. Participants were informed regarding the nature of the task and were explicitly told to take advantage of the cue by preparing the indicated responses. They were instructed to react as quickly and accurately as possible by pressing the correct response key. When pressing an incorrect response key, an error message appeared on the screen. Participants received 20 practice trials at the start of the experiment. The intertrial interval was 1 s.
A
Reaction Time (ms)
positions (uncued condition) or two of the four positions (cued conditions). Then, after a preparation interval of 2, 3, 4, or 5 s, the target stimulus was presented below one of the cued positions. The warning, cue, and target signal remained visible for the duration of the trial.
Reaction Time (ms)
320
RESPONSE PREPARATION IN AGING
A
Reaction Times MIXED
Reaction Time (ms)
800
FIXED
700
Old
600
alent precuing benefits. For older participants, this ANOVA produced a significant main effect of cue type, F(2,52) = 13.50, P < .001. Subsequent planned comparisons revealed significantly shorter RTs for hand cues than for finger and neither cues (P < .001), thus demonstrating a significant hand advantage in older age of 25 ms.
500
Young
400 300
Neither
Finger
Hand
Uncued
Neither
Finger
Hand
Uncued
Cue
Errors MIXED
FIXED
3 2.5 2
Young Old
1.5 1 0.5 0
Neither
Finger
Hand
Uncued
Neither
Finger
Hand
Uncued
Cue Figure 3. (a) Mean reaction time and (b) mean percentage errors of the younger and older participants as a function of cue type and presentation mode (mixed vs. fixed preparation intervals). The dotted horizontal line indicates the mean RT of the control, uncued condition, which provides a baseline to evaluate the RT benefits driven by the hand, finger, and neither cues. The dotted circles highlight the significant advantage for hand cues over finger and neither cues in the group of older participants. The reported means are collapsed over the factor preparation interval. Error bars denote the between-participants standard error of the mean.
interaction, F(3,144) = 3.55, P < .02, indicated greater preparation benefits for fixed than mixed preparation intervals. The preparation interval × presentation mode interaction, F(3,144) = 66.29, P < .001, indicated opposite effects of longer preparation intervals for mixed and fixed conditions: with mixed preparation intervals RT shortened with increasing preparation interval, whereas with fixed presentations RT lengthened with increasing preparation interval. The most important finding, however, was a significant age × cue type interaction, F(3,144) = 3.40, P < .05, which demonstrated that older but not younger participants showed an RT advantage for hand cues (over finger and neither cues). This effect is depicted in Figure 3a and was statistically confirmed by separate ANOVAs for younger and older participants conducted on the RTs for the three informative cue conditions (hand, finger, and neither cued). For younger participants, this ANOVA revealed no significant effect of cue type, F(2,44) < 1, thus demonstrating a pattern of equiv-
Errors The overall mean error rate was 1.40%. The ANOVA on arcsine-transformed error rates demonstrated significant main effects of cue type, F(3,144) = 3.76, P < .05, and preparation interval, F(3,144) = 3.58, P < .05. These main effects indicated that uncued (M = 1.82%), and neither-cued (M = 1.59%) conditions generated more errors than did hand-cued (M = 1.04%) and finger-cued (M = 1.14%) conditions and that the shortest preparation interval of 2 s generated more errors than did the other preparation intervals (M values = 1.67%, 1.39%, 1.17%, and 1.36% for increasing intervals, respectively). The main effect of age was not significant, F(1,48) < 1, but entered in two significant interactions. The preparation interval × age interaction, F(3,144) = 2.86, P < .05, indicated that only younger adults showed an effect of preparation interval (M values = 2.09%, 1.47%, 1.25%, and 1.33% for increasing intervals, respectively), F(3,66) = 2.86, P < .01, older adults did not (M values = 1.25%, 1.32%, 1.09%, and 1.39% for increasing intervals, respectively), F(3,78) = 1.13, P = .34. The age × cue type interaction, F(3,144) = 8.58, P < .001, mirrored the RT data in showing a hand advantage for older participants but not for younger ones (see Figure 3b). That is, younger participants showed a pattern of equal error rates for hand, finger, and neither cues, F(2,44) < 1, whereas older participants showed a pattern of differential error rates, with hand cues producing fewest errors and neither cues most: 0.72%, 1.23%, and 2.04% for hand, finger, and neither cues, respectively, F(2,52) = 10.99, P < .001. This hand advantage in response accuracy in older age was independent of presentation mode, F(3,144) = 1.78, P > .15 (see Figure 3b). The age × cue type interaction also indicated that younger adults made more errors in the uncued condition than in the hand-, finger- and neither-cued conditions (P < .01), whereas older adults made fewer errors in the uncued condition than in the neither-cued condition (P < .01). Discussion This study resolved a discrepancy in the literature concerning the maximal preparation benefit that older adults can attain when preparing two fingers on one as opposed to two hands. In particular, we examined the effect of temporal uncertainty on the efficiency of within- versus betweenhands preparation in aging and reported five key findings. First, we found opposite effects of longer preparation intervals for mixed and fixed designs: with mixed preparation intervals RT shortened with increasing preparation interval,
Downloaded from http://psychsocgerontology.oxfordjournals.org/ by guest on January 7, 2016
B
Errors (%)
321
322
ADAM ET AL.
According to the Grouping Model (Adam, Hommel, & Umiltà, 2003, 2005), within-hands preparation proceeds fast and automatic, whereas between-hands preparation requires slower and effortful top–down processing. This difference in processing efficiency produces a within-hands preparation superiority effect but only with short preparation intervals. This is true for young adults. Older participants, on the other hand, appear to exhibit a structural difference in between-hands preparation that cannot be remedied by extending the preparation interval to 5 s. Perhaps, extending the preparation interval even further might eliminate the between-hands difference, but this seems unlikely given the lack of effects in the present 2- to 5-s range. How to explain the age-related between-hands preparation difference? If the Grouping Model is correct in claiming that within-hands preparation proceeds fast and automatic, whereas between-hands preparation requires slow, effortful top–down processing, the finding of spared within-hands preparation and impaired between-hands preparation in older age accords with the notion that advancing age is accompanied by a reduction in central processing resources while leaving automatic processes intact (e.g., Allen, Groth, Weber, & Madden, 1993; Hasher & Zacks, 1979; Wishart, Lee, Murdoch, & Hodges, 2000). Alternatively, the present findings can be brought in line with two other theories of cognitive aging. First, according to the competing mental set selection hypothesis, age differences are particularly large when competing and potentially confusable mental sets exist in the same performance context (e.g., Mayr & Liebscher, 2001). Supporting evidence comes from studies that used the task-switching paradigm, which requires participants to select on each trial both the responses to be given to a particular stimulus and the mental set that specifies the valid stimulus dimension and stimulus–response rule (e.g., Mayr & Kliegl, 2000; Mayr & Liebscher, 2001). Similarly, in the present finger-cuing task, participants have to respecify on each trial the relevant stimulus–response set based on cue information. According to the mental set selection hypothesis, age differences are particularly large in coordinatively complex conditions. Of course, this was strongly the case in the current betweenhands cue conditions, which required the selection of fingers across instead of within hands and hemispheres. Second, Braver and Barch (2002) advanced a contextprocessing theory, postulating an impaired “context-processing mechanism” in aging. According to this model, flexible and successful performance on a wide variety of cognitive tasks critically depends on the internal representation, maintenance, and updating of context information. In this model, “context” is a broadly defined construct referring to taskrelevant information provided by a cue or task instruction, which enables modulation of input and output processes. Moreover, context information may influence planning processes, particularly in situations where there is strong competition between response alternatives. The finger-cuing task
Downloaded from http://psychsocgerontology.oxfordjournals.org/ by guest on January 7, 2016
whereas with fixed presentations RT lengthened. This finding indicates that our manipulation of temporal uncertainty was effective and in line with previous studies (e.g., Niemi & Näätänen, 1981). Second, young participants prepared res ponses on different hands as effectively as responses on the same hand, regardless of the duration and presentation mode (fixed vs. mixed) of the preparation intervals. This effect occurred for speed (RT) and accuracy (errors) of responding. These outcomes replicate previous reports showing cue equivalence for young adults with preparation intervals of 2 s and longer (e.g., Adam, Hommel, & Umiltà, 2003). Third, older participants consistently showed longer RTs and more errors for cues indicating between-hands preparation than for cues indicating within-hands preparation, revealing an age-related difference specifically for between-hands preparation. Fourth, this effect was independent of temporal uncertainty, refuting the hypothesis that older adults show a difference in between-hands preparation only with fixed preparation intervals. Fifth, the between-hands preparation difference in older age was also independent of the length of the preparation interval, refuting the idea that insufficient preparation time underlies this difference (Proctor et al., 2006). There are now four published studies on the maximal preparation benefit that older people can attain: three studies reported an age-related difference for between-hands preparation (Adam et al., 1998; Proctor et al., 2006; the present study) and one study reported age equivalence (Moresi et al., 2009). The latter study deviated from the first three studies in using a mixed instead of a fixed design, but the present study ruled out the relevance of this factor. Hence, taken together, the available evidence seems to converge on the conclusion that older people exhibit a structural difference in preparing two fingers on two hands but not on one hand, regardless of temporal uncertainty. It is important to rule out an alternative explanation of the preparation advantage for hand cues. According to the “spatial proximity” hypothesis, preparation for two stimulus positions is more efficient the closer together they are, possibly because of an advantage in sharing attention across nearby positions (Miller, 1982). This hypothesis, however, can be rejected because it is not supported by the observed effects. Specifically, the spatial proximity hypothesis would predict shorter RTs when the two index fingers are cued than when the two middle fingers are cued, simply because the cued locations are in closer proximity in the former than in the latter situation. The RT results, however, showed the exact opposite pattern: RTs for preparing the two index fingers in the present study were significantly longer than RTs for preparing the two middle fingers (559 vs. 498 ms), F(1,48) = 66.78, P < .001. Similar findings (e.g., Adam, 1982; Miller, 1982) have been attributed to a greater confusability for inner than for outer stimulus–response elements in a linear array because inner elements have two flanking neighbors, whereas outer elements have only one (Adam, Van Boxtel, Houx, Van Gerven, & Jolles, 2006).
RESPONSE PREPARATION IN AGING
In conclusion, the present paper demonstrates that advancing age impairs the ability to prepare two fingers on different hands but spares the ability to prepare fingers on the same hand, regardless of temporal uncertainty. This empirical clarification is important for two reasons. First, it highlights a structural age-related limit in between-hands preparation that cognitive and neural models of motor behavior in aging should account for. Second, it qualifies recently published human factor guidelines proposing that older people can benefit at least as much as younger adults from advance information (Proctor et al., 2005). The present study suggests that this is indeed the case for cues inducing within-hands preparation, but not for cues inducing between-hands preparation. Correspondence Correspondence should be addressed to Jos J. Adam, PhD, Faculty of Health, Medicine, and Life Sciences, Maastricht University, PO Box 616, 6200 MD Maastricht, The Netherlands. E-mail: jos.adam@ maastrichtuniversity.nl. References Adam, J. J. (1982). The spatial proximity hypothesis of the hand advantage in spatial precuing tasks. Human Movement Science, 11, 641–652. doi:10.1016/0167-9457(92)90034-9. Adam, J. J., Backes, W., Rijcken, J., Hofman, P., Kuipers, H., & Jolles, J. (2003). Rapid visuomotor preparation in the human brain: a functional MRI study. Cognitive Brain Research, 16, 1–10. doi:10.1016/ S0926-6410(02)00204-5. Adam, J. J., Hommel, B., & Umiltà, C. (2003). Preparing for perception and action: (I) The role of grouping in the response-cuing paradigm. Cognitive Psychology, 46, 302–358. doi:10.1016/S0010-0285(02)00516-9. Adam, J. J., Hommel, B., & Umiltà, C. (2005). Preparing for perception and action: (II) Automatic and effortful processes in response cuing. Visual Cognition, 12, 1444–1473. doi:10.1080/13506280444000779. Adam, J. J., Paas, F. G., Teeken, J. C., van Loon, E. M., van Boxtel, M. P., Houx, P. J., & Jolles, J (1998). Effects of age on performance in a fingerprecuing task. Journal of Experimental Psychology: Human Perception and Performance, 24, 870–883. doi:10.1037/0096-1523.24.3.870. Adam, J. J., Van Boxtel, M. P. J., Houx, P. J., Van Gerven, P. W. M., & Jolles, J. (2006). Perceptual and motor factors mediate the bowed spatial position effect in ageing. European Journal of Cognitive Psychology, 18, 673–685. doi:10.1080/09541440500441600. Allen, P. A., Groth, K. E., Weber, T. A., & Madden, D. J. (1993). Influence of response selection and noise similarity on age differences in the redundancy gain. Journal of Gerontology, 48, P189–P198. Bertelson, P. (1967). Time course of preparation. Quarterly Journal of Experimental Psychology, 19, 272–279. doi:10.1080/14640746708400102. Bherer, L., & Belleville, S. (2004). Age-related differences in response preparation. Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 59(2), 66–74. doi:10.1093/geronb/59.2.P66. Braver, T. S., & Barch, D. M. (2002). A theory of cognitive control, aging cognition, and neuromodulation. Neuroscience and Biobehavioral Reviews, 26, 809–817. doi:10.1016/S0149-7634(02)00067-2. Faust, M. E., Balota, D. A., Spieler, D. H., & Ferraro, F. R. (1999). Individual differences in information-processing rate and amount: Implications for group differences in response latency. Psychological Bulletin, 125, 777–799. doi:10.1037/0033-2909.125.6.777. Gottsdanker, R. (1975). The attaining and maintaining of preparation. In P. M. A. Rabbit, & S. Dornic (Eds.), Attention and performance V (pp. 33–49). London: Academic Press. Gottsdanker, R. (1982). Age and simple reaction time. Journal of Gerontology, 37, 342–348.
Downloaded from http://psychsocgerontology.oxfordjournals.org/ by guest on January 7, 2016
qualifies as a task that explicitly assesses context-processing functions because the cue provides on each trial a new predictive context that facilitates performance by directing attention to a particular set of stimuli and responses. As it stands now, however, the context-processing theory cannot easily accommodate the differential effect of age for within- and betweenhands preparatory cues. Elaborations of this theory, therefore, may be needed, perhaps in terms of differential contributions of automatic and controlled context-processing mechanisms. The relevance of the automatic–controlled processing notion is further illustrated by a recent study that demonstrated an age-related difference in within-hands preparation (Sterr & Dean, 2008). This study used short (i.e., 1,300 ms), symbolic (i.e., letter) cues, which could be valid (80% of the trials) or invalid (20% of the trials). These task characteristics likely required controlled-processing mechanisms, which are sensitive to aging. This is in contrast to the present within-hands preparation condition, which relies to a great extent on automatic processing. Hence, withinhands preparation may or may not be spared in older age, depending on the task and the nature of the underlying preparatory processes. The present study did not find an age-related difference in temporal preparation, as has been documented before. Vallesi, McIntosch, and Stuss (2009), for instance, reported that elderly participants showed an increase instead of a dec rease in RT with longer preparation intervals using a mixed design. Procedural differences here too may explain the discrepancy, in particular the use of symbolic target stimuli (square or triangle) in combination with a restricted number and range of preparation intervals (1 and 3 s). We conjecture that the use of symbolic stimuli, which do not have direct, natural associations with finger responses, may require processing mechanisms that are fundamentally different from those involved in tasks that use spatially compatible stimulus–response arrangements. The crucial difference here may again be associated with the automatic–controlled processing distinction (e.g., Kornblum, Hasbroucq, & Osman, 1990). Brain imaging and electrophysiological studies have revealed a network of brain areas involved in motor preparation, including the sensorimotor cortex, the premotor cortex, the cingulate cortex, and the supplementary motor cortex, as well as the basal ganglia, thalamus, and cerebellum (e.g., Adam et al., 2003; Leuthold & Jentzsch, 2001). According to frontal lobe hypothesis of aging (West, 1996), cognitive aging is related to a structural deficit of inhibitory processes that are supported by the prefrontal cortex. Note that inhibitory processes are particularly important for situations characterized by a strong competition during stimulus and response selection, as is the case for the difficult between-hands cues, which require selective activation and inhibition of relevant and irrelevant responses within and across hemispheres. The finding of increased error rates for between-hands preparation in older age fits with this notion.
323
324
ADAM ET AL.
Niemi, P., & Näätänen, R. (1981). Foreperiod and simple reaction time. Psychological Bulletin, 89, 133–162. doi:10.1037/0033-2909. 89.1.133. Proctor, R. W., & Reeve, T. G. (1986). Salient-feature coding operations in spatial precuing tasks. Journal of Experimental Psychology: Human Perception and Performance, 12, 277–285. doi:10.1037/0096-1523.12. 3.277. Proctor, R. W., Vu, K. P., & Pick, D. F. (2005). Aging and response selection in spatial choice tasks. Human Factors, 47, 250–270. doi:10.1518/0018720054679425. Proctor, R. W., Vu, K. P., & Pick, D. F. (2006). A deficit in older adults’ effortful selection of cued responses. Journal of Motor Behavior, 38, 265–284. doi:10.3200/JMBR.38.4.265-284. Reeve, T. G., & Proctor, R. W. (1984). On the advance preparation of discrete finger responses. Journal of Experimental Psychology: Human Perception and Performance, 10, 541–553. doi:10.1037/0096-1523. 10.4.541. Reeve, T. G., & Proctor, R. W. (1990). The salient-features coding principle for spatial-and symbolic-compatibility effects. In R. W. Proctor, & T. G. Reeve (Eds.), Stimulus-response compatibility (pp. 163–180). Amsterdam: North-Holland. Sterr, A., & Dean, P. (2008). Neural correlates of movement preparation in healthy ageing. European Journal of Neuroscience, 27, 254–260. doi:10.1111/j.1460-9568.2007.05975.x. Stilitz, I. (1972). Conditional probability and components of RT in the variable foreperiod experiment. Quarterly Journal of Experimental Psychology, 24, 159–168. doi:10.1080/00335557243000030. Vallesi, A., McIntosh, A. R., & Stuss, D. T. (2009). Temporal preparation in aging: A functional MRI study. Neuropsychologia, 47, 2876–2881. doi:10.1016/j.neuropsychologia.2009.06.013. Welford, A. T. (1977). Motor performance. In J. E. Birren, & K. W. Schaie (Eds.), Handbook of the psychology of aging (pp. 450–496). New York: Van Nostrand Reinhold. West, R. L. (1996). An application of prefrontal cortex function theory to cognitive aging. Psychological Bulletin, 120, 272–292. doi:10.1037/ 0033-2909.120.2.272. Wishart, L. R., Lee, T. D., Murdoch, J. E., & Hodges, N. J. (2000). Effects of aging on automatic and effortful processes in bimanual coordination. Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 55, 85–94. doi:10.1093/geronb/55.2.P85.
Downloaded from http://psychsocgerontology.oxfordjournals.org/ by guest on January 7, 2016
Hasher, L., & Zacks, R. T. (1979). Automatic and effortful processes in memory. Journal of Experimental Psychology: General, 108, 356–388. doi:10.1037/0096-3445.108.3.356. Hick, W. E. (1952). On the rate of gain of information. Quarterly Journal of Experimental Psychology, 4, 11–26. doi:10.1080/17470215208416600. Hyman, R. (1953). Stimulus information as a determinant of reaction time. Journal of Experimental Psychology, 45, 188–196. doi:10.1037/ h0056940. Klapp, S. T., & Jagacinski, R. J. (2011). Gestalt principles in the control of motor action. Psychological Bulletin, 137, 443–462. doi:10.1037/a0022361. Kornblum, S., Hasbroucq, T., & Osman, A. (1990). Dimensional overlap: Cognitive basis for stimulus-response compatibility—A model and taxonomy. Psychological Review, 97, 253–270. doi:10.1037/0033295X.97.2.253. Lahtela, K., Niemi, P., & Kuusela, V. (1985). Adult visual choice-reaction time, age, sex, and preparedness: A test of Welford’s problem in a large population sample. Scandinavian Journal of Psychology, 26, 357–362. doi:10.1111/j.1467-9450.1985.tb01175.x. Leuthold, H., & Jentzsch, I. (2001). Neural correlates of advance movement preparation: A dipole source analysis approach. Cognitive Brain Research, 12, 207–224. doi:10.1016/S0926-6410(01)00052-0. Los, S. A. (1996). On the origin of mixing costs: Exploring information processing in pure and mixed blocks of trials. Acta Psychologica, 94, 145–188. doi:10.1016/0001-6918(95)00050-X. Mayr, U., & Kliegl, R. (2000). Task-switching and long-term memory retrieval. Journal of Experimental Psychology: Learning, Memory, and Cognition, 26, 1124–1140. doi:10.1037/0278-7393.26 .5.1124. Mayr, U., & Liebscher, T. (2001). Is there an age deficit in the selection of mental sets? European Journal of Cognitive Psychology, 13, 47–69. doi:10.1080/09541440042000214. Miller, J. (1982). Discrete versus continuous stage models of human information processing: In search of partial output. Journal of Experimental Psychology: Human Perception and Performance, 8, 273–296. doi:10.1037/0096-1523.8.2.273. Moresi, S., Adam, J. J., Van Gerven, P. W. M., Werrij, B. G., Van Boxtel, M. P. J., & Jolles, J. (2009). Preparing fingers within and between hands: Examining the maximal preparation benefit in older age. European Journal of Cognitive Psychology, 21, 1121–1136. doi:10. 1080/09541440802685573.
