Stop Task After-Effects The Extent of Slowing During the Preparation and Execution of Movement Peter G. Enticott,1 John L. Bradshaw,1 Mark A. Bellgrove,2 Daniel J. Upton,1 and James R.P. Ogloff1 1

School of Psychology, Psychiatry and Psychological Medicine, Monash University, Australia School of Psychology and Queensland Brain Institute, The University of Queensland, Australia

2

Abstract. In the stop task, response time to the go signal is increased when the immediately preceding trial involves the presentation of a stop signal. A recent explanation suggests that these ‘‘after-effects’’ are due to mechanisms that occur prior to the completion of response selection processes, but it is possible that they instead may reflect a slowed motor response (i.e., deliberate slowing after response selection). The participants completed a novel stop task that allows a differentiation between the time taken to prepare a movement (which incorporates response selection processes) and the time taken to execute a movement (i.e., speed of motor response). If mechanisms underlying stop task after-effects occur prior to the completion of response selection processes, then slowing should only occur during movement preparation. Movement preparation and execution time during go trials were analysed according to the characteristics of the preceding trial. Slowing after a stop trial was found during movement preparation time (regardless of inhibition success on that stop trial), and it further increased during this period when the primary task stimulus was repeated. There was also evidence for general after-effects during movement execution time, but no effect of repetition. These findings support the current theoretical accounts that suggest that repetition-based stop task after-effects are attributable to a mechanism that occurs prior to the completion of response selection processes, and also indicate a possible switch to a more conservative response set (as in signal detection theory terms) that results in deliberate slowing of movement. Keywords: stop task, after-effects, motor preparation, repetition priming, control adjustments

The stop task concerns the inhibition of a response after it has been prepared or initiated (Logan, 1994; Logan & Cowan, 1984). While stop task performance is intended to measure behavioural inhibitory control, there are also after-effects associated with the stop task. Broadly speaking, response time to the go signal (trial n) is increased when the immediately preceding trial (trial n 1) was a stop trial (i.e., involved presentation of the stop signal). This has been found both when the preceding trial involves successful inhibition (signal inhibit trial) (e.g., Kramer et al., 1992, cited in Emeric et al., 2007; Logan, 1994; Rieger & Gauggel, 1999) and also when it involves unsuccessful inhibition (signal respond trial) (e.g., Rieger & Gauggel, 1999; Verbruggen, Logan, Liefooghe, & Vandierendonck, 2008). After-effects are further increased when the consecutive trials (i.e., stop then go) share primary task stimulus properties (Rieger & Gauggel, 1999), analogous to the negative priming situation. Verbruggen et al. (2008), however, report a stimulus repetition effect only for signal inhibit trials, and no signal inhibit after-effects where stimuli are not repeated. Multiple mechanisms have been proposed to explain these stop task after-effects. Initially, Rieger and Gauggel (1999) suggested that signal respond after-effects could reflect participant strategy and that signal inhibit after-effects (in which primary task stimulus properties were repeated) could reflect, at least in part, an inhibitory mechanism similar to that encountered in negative priming (Tipper, 2001).  2009 Hogrefe & Huber Publishers

More recently, Verbruggen et al. (2008) conducted a series of experiments and concluded that two mechanisms are involved. Firstly, signal respond after-effects are thought to reflect between-trial control adjustments. According to this model, failed inhibition causes the individuals to increase their response threshold, adopting a more conservative response set, thereby improving the likelihood of successful stopping on the subsequent trial should the stop signal appear. This is somewhat analogous to the accounts of post-error slowing (Hajcak, McDonald, & Simons, 2003; Li, Huang, Constable, & Sinha, 2006; Rabbitt, 1966), and control adjustments are thought to occur prior to the presentation of trial n. Secondly, signal inhibit after-effects are thought to involve repetition-priming effects. This is somewhat similar to negative priming (Tipper, 2001), and Verbruggen et al. cite a role for episodic retrieval (Neill, Valdes, Terry, & Gorfein, 1992). According to episodic retrieval theory, which has been proposed as an explanatory account of the negative priming effect (Neill et al., 1992), a stimulus that is associated with a forbidden or inhibited response is given a ‘‘do not respond’’ tag in memory. Where the next trial involves a go response to that same stimulus, this ‘‘do not respond’’ tag is retrieved and interferes with the (correct) response to the go signal. This interference subsequently results in a slowed response. This has been further supported in a recent study that found long-term stop task after-effects (i.e., in stimuli repeated between 1 and 20 trials Experimental Psychology 2009; Vol. 56(4):247–251 DOI: 10.1027/1618-3169.56.4.247

248

Enticott et al.: Stop Task After-Effects

after a signal inhibit trial), which is consistent with a memory retrieval account (Verbruggen & Logan, 2008). These explanatory accounts of signal inhibit and signal respond after-effects each suggest a mechanism that operates prior to the completion of response selection processes on trial n. This implies that increased response time in trial n occurs during the preparation of the response (i.e., before an individual begins to execute a selected response). An alternative account, however, is that deliberate slowing of motor output underlies at least some aspects of stop task after-effects. For example, after-effects could reflect the deliberate slowing of a motor response or movement to increase the likelihood of successful stopping. The presentation of a stop signal may prompt the individuals to slow their motor response, which in turn provides a greater opportunity for stopping should the stop signal appear. Indeed, such accounts suggest that the presence of negative stimuli can cause subsequent slowing of motor behaviour (Wilkowski & Robinson, 2006), and stop trials may be construed as such. This seems especially relevant to after-effects following unsuccessful inhibition. At this stage, however, researchers have not investigated which component of the response (i.e., movement preparation or movement execution) is affected. The present study investigates stop signal after-effects using a novel task that allows separate analysis of both movement preparation and movement execution; that is, it provides a distinction between (a) the time taken to prepare a motor response (which incorporates response selection processes) and (b) the time taken to execute that motor response (which reflects the speed of motor behaviour following the completion of response selection processes). The account of Verbruggen et al. (2008) suggests that after-effects are attributable to mechanisms that delay response selection processes (i.e., episodic retrieval and between-trial control adjustment); according to this model,

slowing should only occur during the time taken to prepare a movement. In contrast, if at all the after-effects are attributable to a deliberate slowing of motor output to increase the chances of successful inhibition, slowing should occur for movement execution time. Indeed, it is possible that multiple mechanisms are involved and that slowing occurs during both movement preparation and movement execution.

Method Participants The participants were 31 healthy adults (15 males, 24 right handed; age range: 19–51 years, M = 32.26, SD = 11.66). Estimated full-scale IQ (using the National Adult Reading Test; Nelson, 1982) was within the average range (M = 108.41, SD = 5.17).

Procedure The stop task involved a forced-choice response time measure that allows differentiation between movement preparation time and movement execution time (Enticott, Ogloff, & Bradshaw, 2006, 2008). The participants were seated in front of a foam board in which four boxes were embedded (see Figure 1). There were three buttons (left response key, start [central] key and right response key) and three LEDs (left LED, fixation [central] LED and right LED). The participants were instructed to respond, as quickly as possible, to the presentation of a green LED (presented above either the left or the right response key on each trial), but to attempt to stop that response if the green LED switched to red. Thus, a green LED was the go signal, while a red

Figure 1. Stop task apparatus. Dimensions: Button/LED boxes: 65 mm · 65 mm; distance between boxes: 20 mm; response key diameter: 25 mm; LED diameter: 5 mm and distance between the centre of each response key: 85 mm. Experimental Psychology 2009; Vol. 56(4):247–251

 2009 Hogrefe & Huber Publishers

Enticott et al.: Stop Task After-Effects

LED (which replaced the existing go signal on a stop trial) was the stop signal. Each trial began with the onset of the centrally-located fixation light (yellow LED), which acted as a signal for the depression of the start key. The participants used only the index finger of their dominant hand. They held down the start key until the fixation light extinguished (a variable 500– 1,000 ms after depressing the start key), at which time a green LED appeared above either the left or the right response key. The participants were required to lift their finger from the start key and press the response key below the green LED as quickly as possible. Movement preparation time was the time (ms) from the presentation of the green LED to the release of the start button; movement execution time was the time (ms) from the release of the start button to the depression of the response key. Occasionally, the green LED switched to red, thus indicating the stop signal. The participants were instructed to avoid slowing their responding. For each trial, the target LED remained illuminated until a response was recorded or until after 1,500 ms. Trials in which the participant did not continue to depress the start key until the fixation light extinguished were ceased and repeated; thus, for a trial to be considered valid, it was not possible for individuals to remove their finger from the start button prior to the presentation of the target LED. The participants received feedback, via brief (1,000 ms) illumination of the fixation LED, after each trial: Green for a correct response (response to go signal and inhibition of response to stop signal) and red for an incorrect response (including failed inhibition of response to stop signal). There were four stop signal delay (SSD) intervals: 20%, 40%, 60% and 80% of each individual participant’s mean response time (Carter et al., 2003). SSD began with the extinction of the fixation LED. To determine the mean response time, the participants initially completed 20 go trials. They then completed 4 blocks of 72 randomised trials (i.e., 288 trials), with the stop signal appearing in one-third of the trials on each block (i.e., 96 stop signal trials and 24 for each SSD). Each of the four SSD intervals was presented six times per block. SSD intervals were recalculated after each block to accommodate for any changes in mean go signal response time.

Data Analyses Error trials (not including failed inhibition) were excluded from the analyses. There were very few such trials overall (mean commission errors = 0, mean omission errors = 0.08 and mean impulsivity errors [i.e., release of start key prior to extinction of fixation LED] = 2.04). Go trials were categorised according to whether they were preceded by (a) a go trial (i.e., no signal, NS), (b) a trial in which stopping was successful (i.e., signal inhibit, SI) or (c) a trial in which stopping was unsuccessful (i.e., signal respond, SR). Furthermore, we differentiated between trials that involved repetition of the go stimulus location (i.e., left or right) and those that involved no repetition of the go stimulus location. This resulted in six variables: NSrep, NSno rep, SIrep, SIno rep, SRrep and SRno rep. A 3 (signal: No signal, signal  2009 Hogrefe & Huber Publishers

249

inhibit and signal respond) · 2 (repetition: Repetition or no repetition) repeated measures analysis of variance was conducted for both movement preparation time and movement execution time. Planned comparisons were conducted to further investigate main effects and interaction effects.

Results Six participants (four males) with less than five observations in any of the six variables were excluded from analyses (average number of observations following exclusions: NSrep = 73, NSno rep = 63, SIrep = 17, SIno rep = 16, SRrep = 10 and SRno rep = 10). This resulted in a final dataset of 25 participants (mean age: 29.48 [10.41] years). Stop signal reaction time (221 ms) was similar to that commonly reported among healthy adults in the stop task literature (Chambers et al., 2006; Logan, 1994; Morein-Zamir, Nagelkerke, Chua, Franks, & Kingstone, 2004). As expected, probability of responding was greater with each increasing SSD interval (20% mean response time [i.e., movement preparation time and movement execution time combined; MRT] = 0.04, 40% MRT = 0.18, 60% MRT = 0.47 and 80% MRT = 0.83).

Movement Preparation Time For movement preparation time, there was a main effect of signal (following a Greenhouse-Geisser correction), F(1, 35) = 17.43, p < .001 repetition, F(1, 24) = 5.67, p < .05 and a Signal · Repetition interaction, F(2, 48) = 5.15, p < .01. To clarify this interaction we performed a number of planned comparisons, which are presented in Table 1 and are described below. Signal Inhibit Where the primary task stimulus was repeated in trial n and n 1, SIrep (309 ms) was greater than NSrep (271 ms). Where there was no repetition, SIno rep (295 ms) was still greater than NSno rep (275 ms), indicating that SI after-effects occurred regardless of whether repetition was evident.

Table 1. Planned comparisons Contrast

F

Movement preparation time (df = 1, 24) NSrep SIrep 51.67*** NSno rep SIno rep 21.14*** SIrep SIno rep 12.82** NSrep SRrep 19.33** NSrep SRno rep 17.62** SRrep SRno rep 2.59 Movement execution time (df = 1, 24) NS SI 4.53* NS SR 9.02** SI SR 1.88

gp 2 .68 .47 .35 .45 .42 .10 .16 .27 .07

*p < .05, **p < .01, ***p < .001. Experimental Psychology 2009; Vol. 56(4):247–251

250

Enticott et al.: Stop Task After-Effects

320

Rep No rep

310

ms

300 290 280 270 260 NS

SI

SR

Figure 2. After-effects during movement preparation time. However, SIrep was greater than SIno rep, suggesting that there was an after-effect of repetition beyond that found for no repetition. Overall, then, movement preparation time was greater following stop trials in which a response was successfully inhibited, and further increased when the go stimulus in the consecutive trials was repeated. Signal Respond For repetition trials, SRrep (308 ms) was greater than NSrep. SRno rep (300 ms) was also greater than NSno rep. There was no difference between SRrep and SRno rep, indicating that there was no after-effect associated with stimulus repetition in signal respond trials. Thus, movement preparation time was greater following stop trials in which a response was not inhibited, but there was no additional slowing when the go stimulus in the consecutive trials was repeated. These interaction effects of signal inhibit and signal respond movement preparation time are presented in Figure 2.

Movement Execution Time For movement execution time, there was a main effect of signal, F(2, 48) = 5.30, p < .01, but no main effect of repetition, F(1, 24) = 0.12, p > .05 and No signal · Repetition interaction effect (following Greenhouse-Geisser correction), F(2, 38) = 1.63, p > .05. To further investigate this main effect, we compared NS, SI and SR conditions (see Table 1). Movement execution time for NS (255 ms) was less than that for both SI (265 ms) and SR (273 ms). There was no difference between SI and SR.

Discussion The current study investigated stop task after-effects using a novel apparatus that allowed us to separate response time into two components: Movement preparation and movement execution. This allowed us to determine, for the first time, Experimental Psychology 2009; Vol. 56(4):247–251

when this slowing occurred, thus informing as to the likely mechanism/s underlying these after-effects. The most recent theoretical account of stop task after-effects (Verbruggen et al., 2008) suggests separate mechanisms for signal inhibit and signal respond after-effects (i.e., repetition-priming and between-trial control adjustments), but each is thought to operate prior to the completion of response selection processes (and therefore during the preparation rather than execution of a motor response). Consistent with the previous literature, after-effects were found during movement preparation time following both signal inhibit and signal respond trials, and further enhanced when primary task properties on consecutive trials were repeated. During movement execution time, there was evidence for small but significant general (i.e., repetition and nonrepetition combined) after-effects (i.e., following signal inhibit and signal respond), but no effect of repetition. These findings support the view that repetition aftereffects are entirely confined to the period that occurs prior to the completion of response selection processes. It also suggests that nonrepetition after-effects, thought to reflect between-trial control adjustments, largely take place during this preparatory phase. While the general after-effects found during movement execution time were small, they were nonetheless significant, and at this stage we cannot discount the possibility that slowing of a motor response (e.g., Wilkowski & Robinson, 2006) also contributes to stop task after-effects. While largely supporting the account of Verbruggen et al. (2008), particularly in relation to repetition effects, the current findings were not entirely consistent with Verbruggen et al. As also reported by Rieger and Gauggel (1999), after-effects for signal inhibit trials occurred even where the primary task properties were not repeated (although there was further slowing for stimulus repetition, indicating the presence of a repetition priming effect). This could suggest that the participants made between-trial control adjustments for both signal inhibit and signal respond trials, and this may reflect differences in the nature of the stop tasks used across the various studies (e.g., presence of feedback after each trial in the current study, varying method of setting SSDs, specific instructions given to the participants; see Verbruggen et al., 2008 for a discussion of task-related effects). In the current study, release of the central start button during signal inhibit trials may have been viewed by the participant as a ‘‘partial’’ response, and therefore lead to between-trial control adjustments; unfortunately, however, data recording limitations in our stop task preclude a formal analysis of this possibility. Although certainly an issue deserving further research, our results show that repetition based stop task after-effects were consistently limited to the movement preparation component of the response. The current study is limited in the use of a single method for imposing SSD and a relatively small number of observations for the repetition/no-repetition conditions. Furthermore, our research methodology does not provide a test of the specific mechanism/s underlying after-effects, but it does provide support for the notion that this mechanism occurs prior to response selection. It is therefore consistent with a repetition-priming account of signal inhibit after-effects,  2009 Hogrefe & Huber Publishers

Enticott et al.: Stop Task After-Effects

and not inconsistent with the suggestion that between-trial control adjustments underlie signal respond after-effects. The existence of small, general after-effects during movement execution time is consistent with a slowed motor output explanation, and a slowed motor response may comprise a portion of nonrepetition after-effects. It is possible, however, that this finding emerged because of the significant movement execution demands of the task itself (i.e., 85 mm movement from one button to another), and may not occur in stop tasks where, for example, the participants rest their fingers over the response buttons. In summary, this research provides an increased understanding of the nature of after-effects in the stop task; specifically, that slowing largely occurs during the preparation of a response, and only during this preparatory phase where repetition effects are concerned. Further research should continue to evaluate the specific cognitive mechanisms underlying these interesting effects and to determine the extent to which the slowing of movement execution is also involved.

References Carter, J. D., Farrow, M., Silberstein, R. B., Stough, C., Tucker, A., & Pipingas, A. (2003). Assessing inhibitory control: A revised approach to the stop signal task. Journal of Attentional Disorders, 6, 153–161. Chambers, C. D., Bellgrove, M. A., Stokes, M. G., Henderson, T. R., Garavan, H., Robertson, I. H., et al. (2006). Executive ‘‘brake failure’’ following deactivation of human frontal lobe. Journal of Cognitive Neuroscience, 18, 444–455. Emeric, E. E., Brown, J. W., Boucher, L., Carpenter, R. H., Hanes, D. P., Harris, R., et al. (2007). Influence of history on saccade countermanding performance in humans and macaque monkeys. Vision Research, 47, 35–49. Enticott, P. G., Ogloff, J. R. P., & Bradshaw, J. L. (2006). Associations between laboratory measures of executive inhibitory control and self-reported impulsivity. Personality and Individual Differences, 41, 285–294. Enticott, P. G., Ogloff, J. R. P., & Bradshaw, J. L. (2008). Response inhibition and impulsivity in schizophrenia. Psychiatry Research, 157, 251–254. Hajcak, G., McDonald, N., & Simons, R. F. (2003). To err is autonomic: Error-related brain potentials, ANS activity, and post-error compensatory behaviour. Psychophysiology, 40, 895–903. Li, C. R., Huang, C., Constable, R. T., & Sinha, R. (2006). Gender differences in the neural correlates of response inhibition during a stop task. NeuroImage, 32, 1918–1929. Logan, G. D. (1994). On the ability to inhibit thought and action: A users’ guide to the stop-signal paradigm. In D. Dagenbach,

 2009 Hogrefe & Huber Publishers

251

& T. H. Carr (Eds.), Inhibitory processes in attention, memory, and language. San Diego: Academic Press. Logan, G. D., & Cowan, W. B. (1984). On the ability to inhibit thought and action: A theory of an act of control. Psychological Review, 91, 295–327. Morein-Zamir, S., Nagelkerke, P., Chua, R., Franks, I., & Kingstone, A. (2004). Inhibiting prepared and ongoing responses: Is there more than one kind of stopping? Psychonomic Bulletin & Review, 11, 1034–1040. Neill, W. T., Valdes, L. A., Terry, K. M., & Gorfein, D. S. (1992). Persistence of negative priming: II. Evidence for episodic trace retrieval. Journal of Experimental Psychology: Learning, Memory, and Cognition, 18, 993–1000. Nelson, H. E. (1982). National Adult Reading Test (NART) test manual. Windsor, Berkshire: NFER-Nelson. Rabbitt, P. M. A. (1966). Errors and error correction in choiceresponse tasks. Journal of Experimental Psychology, 71, 264–272. Rieger, M., & Gauggel, S. (1999). Inhibitory after-effects in the stop signal paradigm. British Journal of Psychology, 90, 509–518. Tipper, S. P. (2001). Does negative priming reflect inhibitory mechanisms? A review and integration of conflicting views. The Quarterly Journal of Experimental Psychology, 54A, 321–343. Verbruggen, F., & Logan, G. D. (2008). Long-term aftereffects of response inhibition: Memory retrieval, task goals, and cognitive control. Journal of Experimental Psychology: Human Perception and Performance, 34, 1229–1235. Verbruggen, F., Logan, G. D., Liefooghe, B., & Vandierendonck, A. (2008). Short-term aftereffects of response inhibition: Repetition priming or between-trial control adjustments? Journal of Experimental Psychology: Human Perception and Performance, 34, 413–426. Wilkowski, B. M., & Robinson, M. D. (2006). Stopping dead in one’s tracks: Motor inhibition following incidental evaluations. Journal of Experimental Social Psychology, 42, 479–490.

Received March 6, 2008 Revision received June 12, 2008 Accepted June 18, 2008 Peter G. Enticott Alfred Psychiatry Research Centre Level 1, Old Baker Building The Alfred Melbourne Victoria 3004 Australia Tel. +61 3 9076 6594 Fax +61 3 9076 6588 E-mail [email protected]

Experimental Psychology 2009; Vol. 56(4):247–251