Vibro-tactual choice reaction time in a precuing paradigm

ELSEVIER

Human

Vibro-tactual

Movement

Science 16 (1997) 549-565

choice reaction time in a precuing paradigm

Jos J. Adam a,*, David V. Keyson b, Fred G.W.C. Paas ’ a Department of

Movement Sciences, Maastricht University, PO Box 616. 6200 MD Maastricht. The Netherian& b Institute for Perception ResearchlIPO, P.O. Box 513, 5600 MB Eindhoven, The Netherlands ’ Department of Psychology, Maastricht University, PO Box 616, 6200 MD Maastricht, The Netherlands Received

16 April 1997

Abstract

The goal of the present study was to examine the pattern of precuing benefits in a tactual precuing task. In contrast to visual precuing tasks, where visual precues specify a subset of potential finger responses, the tactual precuing task specifies a subset of finger responses tactually, and hence more directly. In Experiment 1 baseline values were established for fourchoice and two-choice tactual conditions. In Experiment 2 these conditions were embedded in a precuing paradigm with tactual precues specifying a subset of responses. Results showed a pattern of tactual precuing benefits that differed dramatically from the usual pattern found for visual cues. That is, the$nger-cued condition produced a significant precuing advantage of 32 ms that was independent of preparation interval and the hand-cued condition showed a small (5 ms) non-significant precuing benefit. We argue that this differential pattern of results for tactual and visual precues is consistent with the view that tactual and visual precues rely in different degrees on stimulus-response translation processes involved in the selection of precued responses. With visual precues, this selection process is rather indirect and requires the involvement of the translation stage mediating between visual and response codes. With tactual precues, this selection process is more direct and possibly mediated by task-specific productions that relate stimuli directly and automatically to specific fingers. Overall, the results support the view that the mediating role of the translation stage is reduced with tactual stimuli. 0 1997 Elsevier Science B.V.

*Corresponding

author.

Tel.: (31) 43 3881389; fax: (31) 43 3670972; e-mail: [email protected].

SO167-9457197/$17.00 0 1997 Elsevier Science B.V. All rights reserved. PIZSOl67-9457(97)0001 1-O

550

J.J. Adam et ul. i HumanMovement Science 16 (1997) 549-565

PsycINFO classijkation:

2320, 2323, 2330, 2340

Keywords: Reaction time; Precuing; Tactual stimulation;

Response selection

1. Vibro-tactual choice reaction time in a precuing paradigm

About four decades ago Leonard (1959) vibrotactually stimulated the fingertips of subjects and found that reaction time did not increase as a function of an increasing number of tactual alternatives (2, 4, and 8 fingers). By manipulating the frequency and amplitude of vibration, Ten Hoopen et al. (1982) showed that this remarkable deviation from the Hick-Hyman Law (Hick, 1952; H yman, 1953) was not inevitable but strongly dependent on the strength of the vibration. They demonstrated that only strong vibrations produce a flat slope between reaction time and number of tactual choices; weak vibrations do not (see also Smith, 1977). Indeed, weak vibrations follow the Hick-Hyman Law and, thus, show a positive relationship between reaction time and the number of tactual choices. This being so, the effectiveness of precue information on tactual choice reaction time can be studied. Generally, a precue reduces the number of stimulus-response alternatives and typically results in shorter reaction times but certain types of precues produce greater reaction time benefits than others. That at least is the central finding in the visual precuing task where a visual precue specifies two out of four possible finger responses (e.g., Miller, 1982; Reeve and Proctor, 1984, 1990). The goal of the present study was to examine the pattern of precuing benefits in a tactual precuing task. Before we elaborate and justify this intent, it seems to be appropriate to discuss the procedural aspects of the precuing task in general and the pattern of differential precuing benefits in the visual domain in particular.

2. Visual precuing task

The visual precuing task is a variant of the movement precuing technique introduced by Rosenbaum (1980) and further developed by Miller (1982). In the visual precuing task, subjects have to respond to spatial-location visual stimuli with discrete finger responses (usually, the index and middle fingers of both hands). The visual display consists of three horizontal rows representing a warning, precue, and target stimulus, respectively. The warning stimulus

J. J. Adam et al. I Human Movement Science 16 (1997) 549-565

551

consists of four plus (+) signs, indicating the four possible stimulus locations; the precue consists of two plus signs, indicating two possible stimulus locations; and the target stimulus consists of one plus sign, indicating the target stimulus location. The temporal order of these three rows is that first the warning stimulus is presented, then, after a constant delay, the precue, and then, after a variable delay, the target stimulus. The variable delay is called the preparation interval as it reflects the amount of time available to selectively prepare the two stimulus-response alternatives indicated by the precue. In other words, the functional significance of the precue is that it transforms the original four-choice reaction task into a two-choice reaction task. At least three precue conditions can be distinguished. In the hand-cued condition, the precue occupies the two left-most or two right-most stimulus positions and signals so that the response has to be made by one of the two fingers on either the left or the right hand. In the finger-cued condition, the precue occupies the two inner or two outer stimulus positions and indicates that the response has to be made by one of the two homologous fingers on different hands. In the uncued condition, the “precue” consists of four plus signs occupying all four stimulus positions and therefore provides no advance information. This condition is the control condition because it leaves the basic, four-choice reaction task unaltered. Because the Hick-Hyman Law states that a two-choice reaction task results in shorter reaction times than a fourchoice reaction task, precue effectiveness is inferred from a significant reaction time advantage for the two-choice precue condition (i.e., hand-cued and finger-cued) over the control, four-choice precue condition (i.e., uncued). ’ Generally, performance in the visual precuing task reveals that, at short preparation intervals (i.e., intervals up to at least 1500 ms), 2 specification of two fingers on the same hand (hand-cued condition) results in greater reaction time benefits than specification of two fingers on different hands (finger-cued condition). In addition, precuing benefits are relatively small at short preparation intervals and increase with longer preparation intervals. According to Reeve and Proctor (1984, 1985, 1990) and Proctor and Reeve (1986, 1988) this pattern of differential precuing benefits is a function of stim’ Often, but not always (Adam, 1994; Miller, 1985) a fourth cue condition is involved: the condition. This cue condition prepares two non-homologous fingers on different hands. ’ Reeve and Proctor (1988) (Experiment 1) found a distinct same-hand advantage of 25 ms ms preparation interval that disappeared when the preparation interval was extended to 3000 were no intermediate preparation intervals it is difficult to tell when exactly the same-hand disappears (somewhere between 1500 and 3000 ms). The 1500 ms figure, therefore, should be “bottom-value” and likely represents an underestimation.

neither-cued for the 1500 ms. As there advantage considered a

552

J.J. Adam et al. I Human Movement

Science 16 (1997) 549-565

ulus-response translation processes or, in other words, of how fast the responses indicated by the visual precue can be determined. Reeve and Proctor provided evidence for this claim by showing that (a) the advantage for the hand-cued condition over the finger-cued condition is apparent primarily at short precuing intervals (i.e., intervals up to 1500 ms), (b) disappears with extensive practice, and (c) with an overlapped placement of hands (i.e., with fingers of both hands alternating on response keys in the order “right index, left middle, right middle, left index”) the usual advantage for the hand-cued condition (two fingers on one hand) switches to an advantage for the neithercued condition (different fingers on different hands). In other words, the hand-cued advantage really is an advantage for the two left-most and the two right-most responses but not for the left or right hand per se. According to Reeve and Proctor (1984, 1985, 1990) this strongly implicates the stimulus-response translation stage as the locus of the ‘hand advantage’.

3. Purpose of study The goal of the present study was to examine the pattern of precuing benefits in a tactual precuing task. In contrast to visual precuing tasks, where precues visually specify a subset of potential finger responses, the tactual precuing task specifies a subset of finger responses tactually, and hence more directly. This presumably minimizes the role of the stimulus-response translation stage. In contrast to the differential pattern of visual precuing benefits, it was therefore hypothesized that tactual precue efictiveness would not, or at least to a lesser extent, vary as a function of precue condition, and, moreover, would remain invariant as a function of preparation interval. This is so, because tactual precues specify the cued subset of responses quickly and directly and hence preclude or minimize the need for any translational process mediating between stimulus and response codes. Note that (tactual) precue effectiveness is defined as the reaction time benefit in the precued condition(s) relative to the uncued condition; in other words, it is a difference score and should not be confused with the absolute reaction time values of the precued and uncued conditions.

4. Experiment 1: Two-choice and four-choice tactual reactions

The goal of Experiment 1 was twofold, the first goal being to determine whether the vibration strength of our vibro-tactual reaction time apparatus

J. J. Adam et al. I Human Movement Science 16 (1997) 549-565

553

(frequency: 100 Hz, amplitude: 300 pm) was such that it would not eliminate the Hick-Hyman Law effect. In other words, the vibro-tactual reaction time apparatus should produce a reliable reaction time difference between the four-choice and two-choice conditions. Such a demonstration is a prerequisite because precue effectiveness is inferred from a significant reaction time benefit in the two-choice (i.e., precued) condition over the four-choice (uncued) condition. Secondly, Experiment 1 was designed to establish baseline (reaction time) values for the four-choice and two-choice conditions in order to allow evaluation of tactual precue effectiveness in subsequent experimentation (Experiment 2). Hence, Experiment 1 employed no precues but presented two-choice and four-choice conditions blockwise.

5. Method 5.1. Subjects

Twelve volunteers from the Institute for Perception Research, six male and six female, with a mean age of 31 years participated in the experiment. They were all right-handed. 5.2. Apparatus

and stimuli

The vibro-tactual reaction time apparatus had four ‘vibrating’ response keys. The response keys were placed at the vertices of a virtual regular trapezium (lower base, 15.7 cm; upper base, 22.7 cm, and the sides, 1.2 cm) to ensure that they would comfortably accommodate the tips of the index and middle fingers of both hands. The circular response keys had diameters of 15 mm, and required a force of 1.35 N to operate (i.e., contact a microswitch). The response keys were connected to four independent vibrators that could each be energized at 100 Hz by a surrounding coil mounted between the poles of a permanent magnet. Vibration amplitude was 300 pm. Stimulus durations, intervals, responses, and response latencies were timed and recorded by an IBM-AT computer. 5.3. Design Five subconditions were employed: a four-choice, two-choice (left), twochoice (right), two-choice (index), and two-choice (middle). These five sub-

554

J.J. Adam et al. I Human Movement Science 16 (1997) 549-565

conditions made up three main conditions: the four-choice condition, the two-choice:within-hand and the two-choice:between-hands conditions. The four-choice condition involved the index and middle fingers of both hands; this condition is similar to the uncued condition in the precuing paradigm. The two-choice conditions either involved a within-hand repertoire or a between-hands repertoire. The two-choice:within-hand condition involved the index and middle fingers of the left hand (two-choice (left)) or the index and middle fingers of the right hand (two-choice (right)); this corresponds to the hand-cued condition. The two-choice:between-hands condition involved the middle fingers of both hands (two-choice (middle)) or the index fingers of both hands (two-choice (index)); this corresponds to the fingercued condition. Subjects were instructed to keep all four fingers on the response keys in all conditions. The different (sub)conditions were presented in separate blocks of 24 trials with subjects being informed at the beginning of each block which subset of fingers would be required. The order of the blocks was counterbalanced. Six practice trials were conducted at the beginning of each block. Subjects participated in two consecutive sessions each employing all conditions. The first session was considered a familiarization session. Only the data from the second session were analyzed and included in the data analysis. 5.4. Procedure Each trial began with an auditory warning signal lasting 500 ms. After an interval of 1.5 s the tactual stimulus was delivered to a finger tip. Subjects were instructed to react as quickly as possible to the target stimulus by pressing the vibrating response key with the finger resting on it. Pressing the response key stopped the vibration. An incorrect response invoked an auditory error-signal lasting 500 ms. An intertrial interval of 1.5 s separated the response in a trial from the beginning of the next trial. Subjects were blindfolded. 5.5. Data analysis

Reaction times less than 150 ms or exceeding 1 s were considered outliers and were excluded from the data analyses (0.17%). Mean correct reaction times and percentages of errors were calculated for each subject as a function of the three main conditions: the four-choice condition, the two-choice:with-

J. J. Adam et al. I Human Movement Science 16 (1997) 549-565

555

in-hand and the two-choice:between-hands conditions. These dependent variables were entered in a one-way repeated measures analysis of variance (ANOVA). In determining significance levels, the Geisser-Greenhouse correction (Keppel, 1982) was used to control violations of homogeneity of variance and covariance. Post-hoc analyses were carried out using Tukey’s honestly significant difference procedure: an CIlevel of 0.05 was employed to determine statistical significance.

6. Results 6.1. Reaction times

Results of the ANOVA demonstrated

a significant effect, F(2,22) = 16.01, (within-hand: M = 315 ms; between-hands: A4 = 293 ms) showed reliably shorter reaction times than the four-choice condition (M = 330 ms). Also, the twochoice:between-hands condition was reliably faster than the two-choice: within-hand condition.

p < 0.001. Post-hoc analyses indicated that the two-choice conditions

4.2. Errors

Overall error rate was low: 1.1%. No systematic differences were observed between the three main conditions, F(2,22) < 1.

7. Discussion The results of Experiment 1 were quite clear. Both two-choice conditions resulted in significantly shorter reaction times than the four-choice condition. Therefore, the present vibro-tactual reaction time apparatus was considered sufficiently sensitive for studying tactual precue effectiveness. The two-choice:within-hand condition showed longer reaction times than the two-choice:between-hands condition. This finding is consistent with the results of other studies showing that standard two-choice reaction times may be longer when the responses are given by the fingers of the same hand (within-hand repertoire) than by the fingers of the two hands (between-hands repertoire) (e.g., Alain et al., 1993; Annett and Annett, 1979; Hasbroucq

556

J. J. Adam et al. I Human Movement Science 16 (1997) 549-565

et al., 1995; Heuer, 1986). 3 This phenomenon was first reported by Kornblum (1965), and may therefore be labelled the Kornblum-effect.

8. Experiment 2: Vibro-tactual precuing task It was hypothesized that tactual precues reduce or minimize the involvement of the translation stage; consequently, tactual precue effectiveness should not vary as a function of precue condition and, moreover, should be independent of preparation interval. This is so, because tactual precues specify a subset of responses quickly and directly, thereby producing a pattern of equivalent precuing benefits that is independent of preparation interval.

9. Method 9.1. Subjects Twenty-four students and staff members, 12 male and 12 female, volunteered for the experiment (female: mean age 36 yr; male: mean age 33 yr). Nineteen were right-handed, three left-handed, and two reported that they were bimanual. None of the subjects had any previous experience with the experimental task. 9.2. Apparatus and stimuli The vibro-tactual reaction time apparatus described in Experiment 1 was used.

3 Note that the between-hands advantage does not always manifest itself. When four fingers, instead of two, are in contact with response keys (i.e., there are two relevant and two irrelevant responses) a pattern of equivalent reaction times might materialize (Reeve and Proctor, 1988, Experiment 1). In a study that employed a similar four-finger placement condition, Alain et al. (1993) found a small advantage (6 ms) of the between-hands pairing over the within-hand pairing early in practice (day 1); later in practice, this advantage increased (i.e., 8 and 12 ms for days 2 and 3, respectively). Unfortunately, Alain et al. did not report whether these between-hand advantages were significant.

J. J. Adam et al. I Human Movement

Science 16 (1997) 549-565

5.51

9.3. Design

Five preparation intervals were employed (50, 100,250, 500, and 1000 ms) and five (sub)cue conditions: uncued, hand-cued (left), hand-cued (right), finger-cued (index), and finger-cued (middle). Each subject received a block of 80 trials for each preparation interval with the order of blocks being counterbalanced. Within a block of 80 trials there were 16 trials for each of the five (sub)cue conditions. The order of these cuing conditions within a block of 80 trials was random. Twelve practice trials were conducted at the beginning of each block. 9.4. Procedure Each trial began with an auditory warning signal lasting 500 ms. The tactual cue signal was then delivered to the finger tips. The cue signal consisted of the vibration of all four response keys (uncued condition) or of only two of the four response keys (hand-cued and finger-cued conditions). After a variable (preparation) interval the vibration stopped, and (following a blank interval of 250 ms) the target stimulus was then presented. Note that this blank interval of 250 ms that separated the precue and target stimulus, provided 250 ms of additional preparation time, thereby extending the efictive preparation interval. The target stimulus consisted of the vibration of one of the response keys indicated by the cue. Subjects were instructed to react as quickly as possible (without making too many errors) to the target stimulus by pressing the appropriate (i.e., vibrating) response key with the finger resting on it. Pressing one of the response keys stopped the vibration. An incorrect response invoked an auditory error-signal lasting 500 ms. An intertrial interval of 2.5 s separated the response in a trial from the beginning of the next trial. 9.5. Data analysis As in Experiment 1, reaction times less than 150 ms or exceding 1 s were considered outliers and were excluded from data analyses (0.71%). Mean correct reaction times and percentages of errors were calculated for each subject as a function of main preparation condition (uncued, hand-cued, finger-cued) and preparation interval. These dependent variables were entered in a 3 (preparation condition) x 5 (preparation interval) repeated measures ANOVA. In determining significance levels, the Geisser-Greenhouse correction

558

J.J. Adam et al. I Human Movement Science 16 (1997) 549-565

(Keppel, 1982) was used to control violations of assumptions of homogeneity of variance and covariance. Post-hoc analyses were carried out using Tukey’s honestly significant difference procedure: an CIlevel of 0.05 was employed to determine statistical significance. The data of one subject were discarded because of an extremely high error rate (52%).

10. Results

10.1. Reaction times The means of the reaction times for the main preparation conditions as a function of preparation interval are given in Fig. 1. An ANOVA yielded significant main effects for preparation interval, F(4,SS) = 13.58, p < 0.001, and preparation condition, F(2,44) = 40.27, p < 0.001, but no interaction, F&176) = 1.35, p > 0.2. The main effect of preparation interval indicated longer reaction times for longer preparation intervals (MS = 361, 363, 397, 404, and 428 ms for preparation intervals of 50, 100, 250, 500, and 1000 ms, respectively). The main effect of preparation condition indicated reliably faster responses for the fin-

500

+ uncued + hand-cued -D finger-cued

300

’

I 250

500

750

Effective Preparation

I 1000

1250

Interval (ms)

Fig. I. Mean reaction time as a function of the (effective) preparation interval for the uncued, hand-cued, and finger-cued conditions. Note that the effective preparation interval reflects the factual cue presentation time (i.e., 50, 100, 250, 500, and 1000 ms) plus 250 ms of additional preparation time provided by the blank interval of 250 ms that separated the precue and target stimulus.

J. J. Adam et al. I Human Movement

Science 16 (1997) 549%56s

559

ger-cued condition (A4 = 371 ms) compared to both the uncued (M = 403 ms) and hand-cued (M = 398 ms) conditions; responses of the latter two conditions did not differ significantly from each other. In summary, these outcomes show that only the finger-cued condition produced a precuing advantage over the control (uncued) condition; the handcued condition did not. Moreover, the precuing benefit for the finger-cued condition was independent of preparation interval. IO.2. Errors Overall error rate was 5.2%. The means of the percentages of errors as a function of preparation interval and preparation condition are shown in Fig. 2. The ANOVA yielded main effects for preparation interval, F(4,88) = 6.79, p < 0.001, and preparation condition, F(2,44) = 10.54, p < 0.001, but no interaction, F(8,176) = 1.54, p > 0.10. The main effect of preparation interval indicated fewer errors at longer preparation intervals (MS = 7.1, 6.9, 5.4, 4.0, and 2.3% for preparation intervals of 50, 100, 250, 500, and 1000 ms, respectively). This finding combined with the observation that longer preparation intervals were also associated with longer reaction times suggests the possibility of a speed-accuracy trade-

I

101

* uncued

I

I

8ig

6-

E 0

t w

4

2-

0’

I

250

1

I

500

750

Effective Preparation

1000

I

1250

Interval (ms)

Fig. 2. Percentage of errors as a function of the (effective) preparation interval for the uncued, hand-cued, and finger-cued conditions. Note that the effective preparation interval reflects the factual cue presentation time (i.e., 50, 100, 250, 500, and 1000 ms) plus 250 ms of additional preparation time provided by the blank interval of 250 ms that separated the precue and target stimulus.

560

J.J. Adam et al. I Humun Movement

Science 16 (1997) 549-565

phenomenon contaminating the effect of preparation interval on reaction time. The main effect of preparation condition indicated that subjects made more errors in the uncued and hand-cued conditions (5.3 and 6.7%, respectively) than in the finger-cued condition (3.5%). The latter outcome precludes the possibility of a speed-accuracy tradeoff phenomenon for the shorter reaction times in the finger-cued condition as this condition also showed the fewest errors. off

11. Discussion

Experiment 2 yielded two important findings regarding precue effectiveness in a tactual precuing task: (1) the finger-cued condition produced a significant precuing advantage of 32 ms that was independent of preparation interval; (2) the hand-cued condition showed a small (5 ms) non-significant precuing benefit. This pattern of results deviates strikingly from the pattern of precuing benefits generally found in visual precuing tasks, where, for short preparation intervals, the hand-cued condition produces the greatest reaction time benefit, and where longer preparation intervals are associated with greater precuing benefits. For example, in Reeve and Proctor’s study (Reeve and Proctor, 1984) the maximal preparation benefit for all preparation conditions was achieved only with the longest preparation interval of 3 s. This differential pattern of results for tactual and visual precues is consistent with the view that tactual and visual precues rely in different degrees on stimulus-response translation processes involved in the selection of precued responses. With visual precues, this selection process is rather indirect and requires the involvement of the translation stage mediating between visual and response codes. With tactual precues, this selection process is more direct and possibly mediated by task-specific productions that relate stimuli directly and automatically to specific fingers (Anderson, 1982, 1987; Proctor and Reeve, 1988). Note that the present results bear on the debate between Miller (1982, 1985) and Reeve and Proctor (1984, 1985, 1990) concerning the locus of the hand-cued advantage with visual cues. Whereas Miller argued that this advantage reflects motoric (response preparation) processes, Reeve and Proctor argued that this advantage arises in the stimulus-response translation (i.e., response selection) stage and therefore reflects a non-motoric effect. Compared with visual precuing studies, the present study used a different in-

J. J. Adam et al. I Human Movement Science 16 (1997) 549-565

561

put or stimulus modality (i.e., tactual) but kept the output or response part the same. In other words, by using tactual instead of visual stimuli while leaving motoric, response-related factors unchanged, we modified the nature of the stimulus-response relationship. Therefore, the present demonstration of a pattern of tactual precuing benefits that differed dramatically from the usual pattern of visual precuing benefits, is more in line with a stimulus-response translation account of differences in (visual) precuing benefits than with a motoric, response preparation account. Having addressed the difference between tactual and visual precuing benefits, it remains to be explained why in Experiment 2 there was a substantial and significant precuing benefit for the finger-cued condition but not for the hand-cued condition. That is, the precuing benefit for the finger-cued condition corresponded closely to the advantage of the two-choice:between-hands condition over the four-choice condition in Experiment 1 (37 and 32 ms, respectively). The hand-cued condition, on the other hand, did not yield a significant precuing benefit similar to the shorter reaction times for the two-choice:within-hand condition relative to the four-choice condition in Experiment 1 (i.e., a significant difference of 15 ms in Experiment 1 versus a non-significant difference of 5 ms in Experiment 2). Three explanations seem possible. First, it could be argued that the preparation interval was too short. This interpretation, however, seems to be unlikely as the reaction time benefit of the finger-cued condition was already reached at the shortest preparation interval of 50 ms, and, if anything, the (non-significant) precuing benefit of the hand-cued condition decreased with longer preparation intervals. This leads to the second possibility, namely that the preparation interval was too long. It is possible that effective preparation of two fingers on one hand takes place initially, but that the outcome of such preparation is difficult to sustain over time. Inspection of Fig. 1 indeed suggests that a small hand-cued precuing benefit of about 10 ms might have materialized for the shortest preparation intervals of 50, 100, and 250 ms, but not for the longer preparation intervals of 500 and 1000 ms. 4 It this context it is important to realize that the blank interval of 250 ms that separated the precue and target stimulus, provided 250 ms of additional preparation time, thereby extending the effective preparation interval. In agreement with this possibility is the fact that the spinal motor structures are more sensitive to central commands at 4 A two-way ANOVA for the hand-cued and uncued conditions over the three shortest preparation intervals yielded a (near)significant main effect of preparation condition, F( 1,22) = 3.87, p = 0.062.

562

J.J. Adam et al. I Human Movement Science 16 (1997) 549-565

short rather than at long intervals (for reviews, see Bonnet et al., 1982; Requin et al., 1991). A third possibility is that, via some kind of interhemispheric transfer process, selective preparation of two fingers on the same hand “spills over” to the other hemisphere thereby priming the corresponding fingers of the other hand. According to this possibility, the net effect of precuing two fingers on one hand would be similar to precuing all four fingers (i.e., the uncued condition). This explanation is similar to the hypothesis that there are functional links between homologous fingers on different hands (Rosenbaum, 1991). That is, neural representations of homologous fingers may be linked so that activation of one neural representation primes the other. In support of this hypothesis, Rosenbaum (199 1) cites several experimental findings. First, when subjects are asked to press two keys at once, homologous finger combinations (e.g., left and right index fingers) are performed more quickly than non-homologous finger combinations (e.g., left index finger and right middle finger) (Rabbitt et al., 1975). Second, in a four-choice reaction time task, the incidence of erroneous homologous responses is much greater than erroneous non-homologous responses (Rabbitt and Vyas, 1970). That is, errors due to an incorrect keypress produced by a homologous finger response occur thrice as frequently than those produced by a non-homologous finger response. Third, research on typewriting has shown that homologous finger substitutions constitute a very common category of typing error (e.g., typing d instead of k with the middle finger). Future research should examine the validity of these conjectures. In Experiment 2, the finger-cued condition showed shorter reaction times than the hand-cued condition. Clearly, this result reflects the Kornblum-effeet, or in other words, that the between-hands choices are faster than the within-hand choices. ’ Two hypotheses have been proposed to account for this effect: the response-preparation hypothesis (Kornblum, 1965; Rosenbaum and Kornblum, 1982) and the response-implementation hypothesis (Reeve and Proctor, 1988; Hasbroucq et al., 1995). According to the response-preparation hypothesis, the between-hands repertoire allows better 5 It is possible to argue that the absence of a ‘between-hand, two-choice condition involving nonhomologous fingers on different hands’ contributes to the between-hand advantage found in the present study. This possibility, however, can be rejected because it is not consistent with the available evidence. That is, Kornblum (1965) demonstrated that the between-hand advantage also holds for non-homologous different-hand couplings. Furthermore, and most importantly, Reeve and Proctor (1988); Experiment I, two-finger constant condition) and Hasbroucq et al. (1989) demonstrated that homologous and nonhomologous finger (between-hand) pairings do not differ in producing a between-hand advantage.

J.J. Adam et al. I Human Movement Science 16 (1997) 549-565

563

preparation and, thus, faster selection of the response than the within-hand

repertoire. To be specific, according to Kornblum (1965), two mechanisms may play a role: (1) response activation, and (2) response interference. The latter operation involves the inhibition or suppression of competing response alternatives. Kornblum (1965) suggested that the degree of response interference generated by a competing response(s) increases as its activation level increases. This proposal of a direct relationship between activation and interference has recently found some experimental support (Buckolz et al., 1994). In contrast, the response-implementation hypothesis contends that, due to neuro-anatomical constraints, response execution is faster in the betweenhands repertoire than in the within-hand repertoire. The present finding that the Kornblum-effect also materializes with tactual stimuli, and, moreover, is found both in a standard two-choice reaction paradigm (Experiment 1) and in a precuing paradigm (Experiment 2) seems to be more in line with the response-implementation hypothesis than with the response-preparation hypothesis of this effect. That is, given that tactual stimulation bypasses or minimizes the stimulus-response translation stage and thereby ensures an extremely efficient response-preparation process, reaction time differences between the within-hand and between-hands repertoires seem to be tied more to differences in response execution than to differences in response preparation. Similarly, the fact that both Experiment 1 (where the two relevant responses were known in advance and remained constant throughout a block of trials) and Experiment 2 (where the two relevant responses changed from trial to trial and received relatively short preparation intervals) showed a significant and substantial Kornblum-effect (22 and 27 ms, respectively), suggests that the process of response execution rather than the process of response preparation constitutes the bottleneck in the Kornblum-effect. This interpretation is consistent with the results of a recent study by Hasbroucq et al. (1995) who, using electromyographic (EMG) recordings, examined the effects of repertoire on the premotor and motor times of the flexor digitorum prufundus and flexor digitorum sublimis. Their results led to rejection of the response-preparation hypothesis and instead supported the view that the central command for the flexion of the right middle finger differs according to the type of repertoire. The command appears to specify a lower rate of recruitment of the prime movers in the within-hand repertoire than in the between-hands repertoire. Finally, it should be noted that the present results and interpretations do not necessarily apply to experimental paradigms that used two instead of

564

J. J. Adam et al. I Human Movement Science 16 (1997) 549-565

four executable responses (see e.g., Reeve and Proctor, 1988; Buckolz et al., 1994).

Acknowledgements We thank Eric Buckolz, Don Bouwhuis, and one anonymous reviewer for valuable comments and suggestions on this work.

References Adam, J.J., 1994. Manipulating the spatial arrangement of stimuli in a precuing task. Acta Psychologica 85, 183-202. Alain, C., Buckolz, E., Taktak, K., 1993. Same-hand and different-hand finger pairings in two-choice reaction time: Presence or absence of response competition? Journal of Motor Behavior 25, 45-5 1. Anderson, J.R., 1982. Acquisition of cognitive skill. Psychological Review 89, 369406. Anderson, J.R., 1987. Skill acquisition: Compilation of weak-methods problem solutions. Psychological Review 94, 192-210. Annett, M., Annett, J., 1979. Individual differences in right and left reaction time. British Journal of Psychology 70, 393404. Bonnet, M., J. Requin, A. Semjen, 1982. Human reflexology and motor preparation. In: D.I. Miller (Ed.), Exercise and Sport Science Reviews. Franklin Institute Press, Philadelphia. Buckolz, E., Stapleton, P., Alain, C., 1994. Activation-interference association for finger responses in choice reaction time tasks. Acta Psychologica 86, l-29. Hasbroucq, T., Guiard, Y., Komblum, S., 1989. The additivity of stimulus-response compatibility with the effects of sensory and motor factors in a tactile choice reaction time task. Ada Psychologica 72, 139-144. Hasbroucq, T., Mouret, I., Seal, J., Akamatsu, M., 1995. Finger pairings in two-choice reaction time tasks: Does the between-hands advantage reflect response preparation? Journal of Motor Behavior 27,251262. Heuer, H., 1986. Intermanual interactions during programming of finger movements. Transient effects of homologuous coupling. In: Heuer, H., Fromm, C., (Eds.), Generation and Modulation of Action Patterns. Springer, Berlin. Hick, W.E., 1952. On the rate of gain of information. Quarterly Journal of Experimental Psychology 4, 1 l-26. Hyman, R., 1953. Stimulus information as a determinant of reaction time. Journal of Experimental Psychology 45, 188-196. Keppel, G., 1982. Design and Analysis. Prentice-Hall, Englewood Cliffs. Komblum, S., 1965. Response competition and/or inhibition in two-choice reaction time. Psychonomic Science 2, 55-56. Leonard, J.A., 1959. Tactual reactions: I. Quarterly Journal of Experimental Psychology 11, 7683. Miller, J., 1982. Discrete versus continuous models of human information processing: In search of partial output. Journal of Experimental Psychology: Human Perception and Performance 8, 273-296. Miller, J., 1985. A hand advantage in preparation of simple keypress responses: Reply to Reeve and Proctor (1984). Journal of Experimental Psychology: Human Perception and Performance 11, 221233.

J. J. Adam et al. I Human Movement

Science 16 (1997)

549-565

565

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. Proctor, R.W., Reeve, T.G., 1988. The acquisition of task-specific productions and modification of declarative representations in spatial precuing tasks. Journal of Experimental Psychology: General 117. 182-196. Rabbitt, P.M.A., Feamley, S., Vyas, SM., 1975. Programming sequences of complex responses. In: Rabbitt, P.M.A., Dornic, S. (Eds.), Attention and performance V. Academic Press, New York. Rabbitt, P.M.A., Vyas, S.M., 1970. An elementary preliminary taxonomy for some errors in laboratory choice RT tasks. Acta Psychologica 33, 5676. 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. Reeve, T.G., Proctor, R.W. 1985. Non-motoric translation processes in the preparation of discrete finger responses: A rebuttal of Miller’s (1985) analysis. Journal of Experimental Psychology: Human Perception and Performance 11, 234240. Reeve, T.G., Proctor, R.W., 1988. Determinants of two-choice reaction-time patterns for same-hand and different-hand finger pairings. Journal of Motor Behavior 20, 317-340. Reeve, T.G., Proctor, R.W., 1990. The salient-features coding principle for spatial- and symboliccompatibility effects. In: Proctor, R.W., Reeve, T.G. (Eds.), Stimulus-Response Compatibility. NorthHolland, Amsterdam. Requin, J., Brener, J., Ring, C., 1991. Preparation for action. In: Jennings, J.R., Coles, M.G.H., (Eds.), Handbook of Cognitive Psychophysiology: Central and Autonomic Nervous System Approaches. Wiley, New York. Rosenbaum, D.A., 1980. Human movement initiation: Specification of arm, direction, and extent. Journal of Experimental Psychology: General 109, 444474. Rosenbaum, D.A., 1991. Human Motor Control. Academic Press, New York. Rosenbaum, D.A., Komblum, S., 1982. A priming method for investigating the selection of motor responses. Acta Psychologica 51, 223-243. Smith, G.A., 1977. Studies of compatibility and a new model of choice reaction time. In: Dornic, S. (Ed.), Attention and Performance VI. Erlbaum, Hillsdale. Ten Hoopen, G., Akerboom, S., Raaymakers, E., 1982. Vibrotactual choice reaction time, tactile receptor systems and ideomotor compatibility. Acta Psychologica 50, 143-157.