Exp Brain Res DOI 10.1007/s00221-014-3959-0
Research Article
Motor-induced visual motion: hand movements driving visual motion perception Mirjam Keetels · Jeroen J. Stekelenburg
Received: 27 January 2014 / Accepted: 9 April 2014 © Springer-Verlag Berlin Heidelberg 2014
Abstract Visual perception can be changed by cooccurring input from other sensory modalities. Here, we explored how self-generated finger movements (left–right or up–down key presses) affect visual motion perception. In Experiment 1, motion perception of a blinking bar was shifted in the direction of co-occurring hand motor movements, indicative of motor-induced visual motion (MIVM). In Experiment 2, moving and static blinking bars were combined with either directional moving or stationary hand motor movements. Results showed that the directional component in the hand movement was crucial for MIVM as stationary motor movements even declined visual motion perception. In Experiment 3, the role of response bias was excluded in a two-alternative forced-choice task that ruled out the effect of response strategies. All three experiments demonstrated that alternating key presses (either horizontally or vertically aligned) induce illusory visual motion and that stationary motor movements (without a vertical or horizontal direction) induce the opposite effect, namely a decline in visual motion (more static) perception. Keywords Visual motion perception · Motor visual · Multisensory · Visual ambiguity · Motor action
Introduction Visual motion perception is a crucial ability for interacting with the world. For example, when walking through the city, our brains are bombarded with incoming motion M. Keetels (*) · J. J. Stekelenburg Department of Cognitive Neuropsychology, Tilburg University, Tilburg, The Netherlands e-mail: [email protected]
signals of passing cars, bikers, pedestrians, self-movement, etc. Although visual motion perception “seems” a task exclusively of the visual system, in the last decades, many studies have shown that in order to correctly perceive the visual world around us, we integrate input from all cooccurring sensory and motor information as well (Calvert et al. 2004). In the present study, we focus on one specific area of visual perception, namely visual motion perception, and how it is modulated by co-occurring self-generated motor movements. Thus, to what extent do hand movements change the percept of visual motion? Experimental psychology has a long history in investigating the modulatory effects of sensorimotor input on visual perception in general. Most demonstrations have shown the effects of sensorimotor input on visual perception by using bi-stable visual stimuli. Bruno et al. (2007), for example, showed that motor exploration reduces visual ambiguity in the Necker cube. When participants touched or actively explored an actual Necker cube, both the number and the durations of illusory percepts were changed (Bertamini et al. 2010; see for comparable results with the Ames’ window: Bruno et al. 2006; see also: Ando and Ashida 2003). Also, in a binocular rivalry study, Maruya et al. (2007) showed that a hand movement that matches one of the visual alternatives helps to disambiguate visual perception. Participants dichotically viewed a flickering grating and a rotating sphere whose rotation was, on congruent trials, under control of the participant’s hand movements. Dominance periods of the congruently rotating visual stimulus were prolonged, and suppression periods were shortened (see also Lunghi and Alais 2013; Klink et al. 2012; Maruya et al. 2007; Lunghi et al. 2010). Taken together, these studies are typical examples of studies showing how exploratory motor actions can modulate the perception of bi-stable static figures toward one or the other percept.
13
One of the first demonstrations in which the effects of directional motor actions on visual motion perception were explored was by Wohlschläger (2000). In this study, participants were presented with an ambiguous circular visual motion display that could either be perceived as rotating clockwise or rotating counterclockwise. Slightly before (i.e., 280 ms) the onset of the visual motion, the participant started a knob-turning action in a clockwise or counterclockwise direction. The results showed that the direction of the performed motor action primed the direction of the perceived motion (called “assimilative influences of action on concurrent perception” by Schütz-Bosbach and Prinz 2007). So, when participants made a clockwise hand rotation, more clockwise visual motion was seen and when a counterclockwise hand rotation was made, more counterclockwise percepts were measured. Similar results were found when participants performed a key-press action instead of a rotational knob-turning action, or when the participant only planned to make the hand movement (but did not actually perform it). No motion bias was found though when the motor action and visual motion rotated on different axes (rotation around a vertical vs. horizontal axis). Wohlschläger (2000) concluded that priming of rotational visual motion occurred when the action and visual motion at least share a common cognitively specified dimension. Comparable findings were demonstrated in a study by Beets et al. (2010). In their study, participants saw an ambiguous structure-from-motion cylinder (400 dots in the shape of a rotating cylinder) that was alternatingly perceived as rotating clockwise or counterclockwise. While viewing this bi-stable visual figure, participants had to make motor movements (i.e., rotating a manipulandum) that are either dependent on their perceptual state (congruent or incongruent) or predefined and independent on their perceptual state (clockwise or counterclockwise). The dependent measure in their study was the dominance durations (i.e., how long does an observer stay in one perceptual state). The results showed that action affects perception, but only when the performed movement is dependent on the current percept. When the action was performed in a predefined direction, it did not change the dominance durations. One could question though whether the null effect in the predefined motor condition is caused by the independency between the visual and motor motion or whether participants might have adopted the strategy to ignore the predefined—constant in one direction—motor movements. An important question that arises from studies examining the effect of motor actions on visual motion perception is about the origin of the effect. On the one hand, the reported effects of motor actions affecting visual motion perception can be caused by the interaction between action
13
Exp Brain Res
and perception, in such a way that the motor action genuinely changes the visual percept of the motion (i.e., perceptual explanation). On the other hand, it might be that response strategies were changed according to the performed motor actions in such a way that when the visual percept is ambiguous while performing a directional hand movement, the participant is more likely to “report” the direction of the motor movement (i.e., strategy bias explanation). In other words, is the motor perception interaction a genuine perceptual effect, or does the co-occurring motor action change the participant’s response strategy? In three experiments, we explored the effect of active motor movements of the hand on visual motion perception. In Experiments 1 and 2, we replicated the original findings of motor actions biasing visual motion perception, and in Experiment 3, we tested the perceptual origin of the effect by using an experimental design that completely excludes response strategies. Experimental paradigms were adopted from the series of studies by Hidaka et al., in which sound-induced visual motion was demonstrated (SIVM; Hidaka et al. 2009; Teramoto et al. 2010b, 2012; Hidaka et al. 2011). In Experiment 1, a moving visual bar was combined with hand motor movements in the same or opposite direction. In Experiment 2, moving or static bars were combined with moving or stationary motor actions in order to examine the necessity of the directional component of the motor action, and whether a stationary motor action can also induce the opposite effect (less motion perception). In Experiment 3, a response strategy bias explanation was excluded by the use of a discrimination task, in which subsequently two moving bars were presented moving in opposite directions, while being combined with either moving or stationary motor actions. The results of all three experiments demonstrate that alternating key presses (either horizontally or vertically aligned) drive illusory visual motion.
Experiment 1, up–down In Experiment 1, participants judged the direction of a blinking horizontal bar that either jumped upward or downward while participants made two co-occurring finger movements in either upward (lower button followed by upper button) or downward (up–down) fashion. The button presses triggered the presentation of the bar either congruently (e.g., up–down bar motion and up–down button pressing) or incongruently (e.g., up–down bar motion and down–up button pressing). If the direction of the motor movement affects visual motion perception, then a static bar (or one that is slightly moving) will be perceived as “upward” when a participant makes an upward motor
Exp Brain Res
movement, while it will be perceived as “downward” when it is combined with a downward motor movement.
Methods Participants Fifteen students (all female; mean age: 18.9 years; 10 righthanded) from Tilburg University participated and received course credits for their participation. Participants reported normal hearing and touch and normal or corrected-to-normal seeing. They were tested individually and were unaware of the purpose of the experiment. Written informed consent was obtained from each participant.
Stimuli Participants sat at a table in a dimly lit and sound-attenuated booth, while resting their head on a chin and forehead rest. Participants rested their right hand on the table with their thumb and index finger on the lowest (closest) and highest (farthest) button of a response box, respectively (total of 5 buttons). See Fig. 1a, b for a schematic view of the experimental setup. Visual stimuli were presented on a CRT display (refresh rate of 100 Hz) that was seen via a mirror in such a way that the perceived location of the visual stimuli approximately matched the location of the motor movements (distance eyes–mirror–screen was 60 cm). Visual stimuli were presented on a black background. A red fixation dot (radius .15°) was located on the left side of the screen (8° left of
Fig. 1 a One trial consists of a visual bar movement either up or down (random). TASK: bar up or down? Motor movements are up or down (random between trials). b Example of trial in which the motor movent is in opposite direction of the visual movement. Result: “Nulling” (visual direction neutralized). c PSE values on the different motor-movement conditions. PSE values represent the bar motion in degrees at which the participant perceives the bar as static (a positive PSE value represents more downward responses)
13
center). The visual stimulus consisted of a white horizontal bar (length: 2.5°; width: .15°) located at either 4 or 14° to the right of the fixation point. A close and far condition was used in order to examine the effect of eccentricity as has been demonstrated for the SIVM effect (Hidaka et al. 2009). Vertical bar movement was induced by flashing two bars successively for 200 ms with a 600 ms ISI while vertical locations were slightly different. Vertical movement was either .075°, .225°, .375°, or .525° up or down (e.g., in the .525° upward condition, the first bar was presented .26° below the screen midpoint and the second bar .26° above screen midpoint). Auditory pacing signals were presented via headphones (3-ms duration; 75 dB when measured at .5 cm from the headphone), and in order to mask the sound of pressing the response box keys, continuous background noise was presented by two loudspeakers on the left and right side of the monitor (61 dB measured at head location). Participants were instructed to keep fixation on the fixation dot and were monitored by the experimenter via a CCTV system (IR Camera–Conrad CT2100C) that was adjusted in such a way that pupil directions could be followed. Design Three within-subject factors were used: bar motion (−.525°, −.375°, −.225°, −.075°, .075°, .225°, .375°, .525° upward; eight steps of .15° with negative numbers indicating a downward movement and positive numbers an upward movement), eccentricity (close or far, respectively, 4° or 14° from fixation), and motor movement (upwards, downwards, no motor). The no-motor condition served as a baseline. Bar motion and eccentricity were varied randomly within blocks of trials, and motor movement was varied between blocks. These factors yielded 48 conditions, of which the far eccentricity was presented 18 times, and the close eccentricity 6 times for a total of 576 trials (432 far and 144 close; the close condition was presented less frequently because it was considered to be an easy condition). Trials were presented in 18 blocks of 32 trials each (6 blocks for each motor-movement condition). At the start of each block of trials, a motor-movement instruction was presented on the screen, either “Up–Down”, “Down–Up” or “Do nothing” (no-motor condition). Procedure Each trial consisted of (1) an auditory pacing signal, (2) a flashing bar with or without motor movement (i.e., the latter in the no-motor condition), and (3) the participant’s upor downward response. 1. At the start of each trial, a blue fixation dot at the left side of the screen was shown and immediately the
13
Exp Brain Res
auditory pacing signal was presented (consisting of 2 clicks with an ISI of 600 ms). The pacing signal indicated the pace at which the participants were required to press the buttons during their motor movement afterward. The fixation dot changed from blue to red 1,500 ms after the second click. 2. In the no-motor blocks, the presentation of two white horizontal bars (200-ms duration with an ISI of 600 ms) followed 400 ms after the color change of the fixation dot. In the motor upward and downward blocks, the fixation color change is the sign for the participant to start their motor movements. In the upward block, the participant pressed first the lower and then the upper button in approximately the same rhythm as the preceding auditory pacing signal (600 ms ISI). Simultaneously, with each key press, the white horizontal bar flashed for 200 ms. In the downward block, first the upper button was pressed and then the lower button. The vertical location of each bar depended on the bar-motion condition (either jumping down–up or up–down with a particular vertical displacement), while the eccentricity depended on the eccentricity condition (close or far from fixation). If the participant pressed the two buttons too fast (tapping interval 700 ms), the participant received feedback on the screen (“too fast” or “too slow”) in order to adjust their inter-tapping interval in subsequent trials. 3. After the bar flashed 2 times, the participant’s task was to judge whether the bar had moved upward or downward by using two dedicated keys with their left hand on the keyboard. The next trial started 1,000 ms after a response was detected. To acquaint participants with the task, experimental blocks were preceded by a practice session. The practice session consisted of 6 blocks (each motor-movement instruction twice) of 32 trials. At the end of the practice session, all participants were accustomed to make the motor movements in the correct rhythm.
Results Trials of the practice session were excluded from the analyses. The average interval between key presses was 591 ms, which indicates that participants performed the motor movements in the correct rhythm. For each participant, the proportion of “upward” responses was calculated for each combination of bar motion (−.525°, −.375°, −.225°, −.075°, .075°, .225°, .375°, .525° upward), eccentricity (close, far), and motor movement (upward, downward, no motor). For each combination of motor movement and
Exp Brain Res Table 1 Average PSEs and JND values for the different motor-movement conditions at close and far eccentricities (SEM between parentheses) Eccentricity
Motor movement PSE
JND
Close to fixation (easy)
Upward Downward No motor Upward
.07 (.00) .07 (.00) .08 (.00) .16 (.02)
Far from fixation (difficult)
Downward No motor
−.07 (.02) .01 (.01) .02 (.02) −.20 (.05) .11 (.03)
.17 (.02)
−.06 (.03)
.20 (.03)
The PSE represents the bar motion in degrees at which the participant perceives the bar as static. The JND represents the displacement in degrees at which 75 % of the responses is correct
eccentricity, an individually determined psychometric function was calculated by fitting a cumulative normal distribution over the eight bar-motion data points using maximum likelihood estimation. One participant was eliminated from the analysis because no reliable psychometric function could be estimated (this participant had a rigid bias toward downward responses even when the bar motion was large and clearly perceivable for other participants; N = 14 remained). The mean of the resulting distribution (the interpolated 50 % crossover point) is the point of subjective equality (henceforth PSE), which represents the bar motion in degrees at which the participant perceives the bar as static (50 % up and 50 % down responses). Besides the PSE, we calculate the just noticeable difference (henceforth JND), which is inversely related to the distribution’s slope and is a measure of a participant’s sensitivity in distinguishing visual upward and downward bar motions. The JND represents the displacement in degrees at which 75 % of the responses is correct. The average PSEs and the JNDs are shown in Table 1 (and see Fig. 1c for a graphical representation of the PSEs). A repeated measures overall ANOVA with motor movement (upward, downward, no motor) and eccentricity (close, far) as within-subject factors was performed on the PSEs and JNDs. JND. The ANOVA on the JND data revealed no main effect of motor movement, F(2,12) = 1.05, p = .37, and no significant interaction effect, F
Research Article
Motor-induced visual motion: hand movements driving visual motion perception Mirjam Keetels · Jeroen J. Stekelenburg
Received: 27 January 2014 / Accepted: 9 April 2014 © Springer-Verlag Berlin Heidelberg 2014
Abstract Visual perception can be changed by cooccurring input from other sensory modalities. Here, we explored how self-generated finger movements (left–right or up–down key presses) affect visual motion perception. In Experiment 1, motion perception of a blinking bar was shifted in the direction of co-occurring hand motor movements, indicative of motor-induced visual motion (MIVM). In Experiment 2, moving and static blinking bars were combined with either directional moving or stationary hand motor movements. Results showed that the directional component in the hand movement was crucial for MIVM as stationary motor movements even declined visual motion perception. In Experiment 3, the role of response bias was excluded in a two-alternative forced-choice task that ruled out the effect of response strategies. All three experiments demonstrated that alternating key presses (either horizontally or vertically aligned) induce illusory visual motion and that stationary motor movements (without a vertical or horizontal direction) induce the opposite effect, namely a decline in visual motion (more static) perception. Keywords Visual motion perception · Motor visual · Multisensory · Visual ambiguity · Motor action
Introduction Visual motion perception is a crucial ability for interacting with the world. For example, when walking through the city, our brains are bombarded with incoming motion M. Keetels (*) · J. J. Stekelenburg Department of Cognitive Neuropsychology, Tilburg University, Tilburg, The Netherlands e-mail: [email protected]
signals of passing cars, bikers, pedestrians, self-movement, etc. Although visual motion perception “seems” a task exclusively of the visual system, in the last decades, many studies have shown that in order to correctly perceive the visual world around us, we integrate input from all cooccurring sensory and motor information as well (Calvert et al. 2004). In the present study, we focus on one specific area of visual perception, namely visual motion perception, and how it is modulated by co-occurring self-generated motor movements. Thus, to what extent do hand movements change the percept of visual motion? Experimental psychology has a long history in investigating the modulatory effects of sensorimotor input on visual perception in general. Most demonstrations have shown the effects of sensorimotor input on visual perception by using bi-stable visual stimuli. Bruno et al. (2007), for example, showed that motor exploration reduces visual ambiguity in the Necker cube. When participants touched or actively explored an actual Necker cube, both the number and the durations of illusory percepts were changed (Bertamini et al. 2010; see for comparable results with the Ames’ window: Bruno et al. 2006; see also: Ando and Ashida 2003). Also, in a binocular rivalry study, Maruya et al. (2007) showed that a hand movement that matches one of the visual alternatives helps to disambiguate visual perception. Participants dichotically viewed a flickering grating and a rotating sphere whose rotation was, on congruent trials, under control of the participant’s hand movements. Dominance periods of the congruently rotating visual stimulus were prolonged, and suppression periods were shortened (see also Lunghi and Alais 2013; Klink et al. 2012; Maruya et al. 2007; Lunghi et al. 2010). Taken together, these studies are typical examples of studies showing how exploratory motor actions can modulate the perception of bi-stable static figures toward one or the other percept.
13
One of the first demonstrations in which the effects of directional motor actions on visual motion perception were explored was by Wohlschläger (2000). In this study, participants were presented with an ambiguous circular visual motion display that could either be perceived as rotating clockwise or rotating counterclockwise. Slightly before (i.e., 280 ms) the onset of the visual motion, the participant started a knob-turning action in a clockwise or counterclockwise direction. The results showed that the direction of the performed motor action primed the direction of the perceived motion (called “assimilative influences of action on concurrent perception” by Schütz-Bosbach and Prinz 2007). So, when participants made a clockwise hand rotation, more clockwise visual motion was seen and when a counterclockwise hand rotation was made, more counterclockwise percepts were measured. Similar results were found when participants performed a key-press action instead of a rotational knob-turning action, or when the participant only planned to make the hand movement (but did not actually perform it). No motion bias was found though when the motor action and visual motion rotated on different axes (rotation around a vertical vs. horizontal axis). Wohlschläger (2000) concluded that priming of rotational visual motion occurred when the action and visual motion at least share a common cognitively specified dimension. Comparable findings were demonstrated in a study by Beets et al. (2010). In their study, participants saw an ambiguous structure-from-motion cylinder (400 dots in the shape of a rotating cylinder) that was alternatingly perceived as rotating clockwise or counterclockwise. While viewing this bi-stable visual figure, participants had to make motor movements (i.e., rotating a manipulandum) that are either dependent on their perceptual state (congruent or incongruent) or predefined and independent on their perceptual state (clockwise or counterclockwise). The dependent measure in their study was the dominance durations (i.e., how long does an observer stay in one perceptual state). The results showed that action affects perception, but only when the performed movement is dependent on the current percept. When the action was performed in a predefined direction, it did not change the dominance durations. One could question though whether the null effect in the predefined motor condition is caused by the independency between the visual and motor motion or whether participants might have adopted the strategy to ignore the predefined—constant in one direction—motor movements. An important question that arises from studies examining the effect of motor actions on visual motion perception is about the origin of the effect. On the one hand, the reported effects of motor actions affecting visual motion perception can be caused by the interaction between action
13
Exp Brain Res
and perception, in such a way that the motor action genuinely changes the visual percept of the motion (i.e., perceptual explanation). On the other hand, it might be that response strategies were changed according to the performed motor actions in such a way that when the visual percept is ambiguous while performing a directional hand movement, the participant is more likely to “report” the direction of the motor movement (i.e., strategy bias explanation). In other words, is the motor perception interaction a genuine perceptual effect, or does the co-occurring motor action change the participant’s response strategy? In three experiments, we explored the effect of active motor movements of the hand on visual motion perception. In Experiments 1 and 2, we replicated the original findings of motor actions biasing visual motion perception, and in Experiment 3, we tested the perceptual origin of the effect by using an experimental design that completely excludes response strategies. Experimental paradigms were adopted from the series of studies by Hidaka et al., in which sound-induced visual motion was demonstrated (SIVM; Hidaka et al. 2009; Teramoto et al. 2010b, 2012; Hidaka et al. 2011). In Experiment 1, a moving visual bar was combined with hand motor movements in the same or opposite direction. In Experiment 2, moving or static bars were combined with moving or stationary motor actions in order to examine the necessity of the directional component of the motor action, and whether a stationary motor action can also induce the opposite effect (less motion perception). In Experiment 3, a response strategy bias explanation was excluded by the use of a discrimination task, in which subsequently two moving bars were presented moving in opposite directions, while being combined with either moving or stationary motor actions. The results of all three experiments demonstrate that alternating key presses (either horizontally or vertically aligned) drive illusory visual motion.
Experiment 1, up–down In Experiment 1, participants judged the direction of a blinking horizontal bar that either jumped upward or downward while participants made two co-occurring finger movements in either upward (lower button followed by upper button) or downward (up–down) fashion. The button presses triggered the presentation of the bar either congruently (e.g., up–down bar motion and up–down button pressing) or incongruently (e.g., up–down bar motion and down–up button pressing). If the direction of the motor movement affects visual motion perception, then a static bar (or one that is slightly moving) will be perceived as “upward” when a participant makes an upward motor
Exp Brain Res
movement, while it will be perceived as “downward” when it is combined with a downward motor movement.
Methods Participants Fifteen students (all female; mean age: 18.9 years; 10 righthanded) from Tilburg University participated and received course credits for their participation. Participants reported normal hearing and touch and normal or corrected-to-normal seeing. They were tested individually and were unaware of the purpose of the experiment. Written informed consent was obtained from each participant.
Stimuli Participants sat at a table in a dimly lit and sound-attenuated booth, while resting their head on a chin and forehead rest. Participants rested their right hand on the table with their thumb and index finger on the lowest (closest) and highest (farthest) button of a response box, respectively (total of 5 buttons). See Fig. 1a, b for a schematic view of the experimental setup. Visual stimuli were presented on a CRT display (refresh rate of 100 Hz) that was seen via a mirror in such a way that the perceived location of the visual stimuli approximately matched the location of the motor movements (distance eyes–mirror–screen was 60 cm). Visual stimuli were presented on a black background. A red fixation dot (radius .15°) was located on the left side of the screen (8° left of
Fig. 1 a One trial consists of a visual bar movement either up or down (random). TASK: bar up or down? Motor movements are up or down (random between trials). b Example of trial in which the motor movent is in opposite direction of the visual movement. Result: “Nulling” (visual direction neutralized). c PSE values on the different motor-movement conditions. PSE values represent the bar motion in degrees at which the participant perceives the bar as static (a positive PSE value represents more downward responses)
13
center). The visual stimulus consisted of a white horizontal bar (length: 2.5°; width: .15°) located at either 4 or 14° to the right of the fixation point. A close and far condition was used in order to examine the effect of eccentricity as has been demonstrated for the SIVM effect (Hidaka et al. 2009). Vertical bar movement was induced by flashing two bars successively for 200 ms with a 600 ms ISI while vertical locations were slightly different. Vertical movement was either .075°, .225°, .375°, or .525° up or down (e.g., in the .525° upward condition, the first bar was presented .26° below the screen midpoint and the second bar .26° above screen midpoint). Auditory pacing signals were presented via headphones (3-ms duration; 75 dB when measured at .5 cm from the headphone), and in order to mask the sound of pressing the response box keys, continuous background noise was presented by two loudspeakers on the left and right side of the monitor (61 dB measured at head location). Participants were instructed to keep fixation on the fixation dot and were monitored by the experimenter via a CCTV system (IR Camera–Conrad CT2100C) that was adjusted in such a way that pupil directions could be followed. Design Three within-subject factors were used: bar motion (−.525°, −.375°, −.225°, −.075°, .075°, .225°, .375°, .525° upward; eight steps of .15° with negative numbers indicating a downward movement and positive numbers an upward movement), eccentricity (close or far, respectively, 4° or 14° from fixation), and motor movement (upwards, downwards, no motor). The no-motor condition served as a baseline. Bar motion and eccentricity were varied randomly within blocks of trials, and motor movement was varied between blocks. These factors yielded 48 conditions, of which the far eccentricity was presented 18 times, and the close eccentricity 6 times for a total of 576 trials (432 far and 144 close; the close condition was presented less frequently because it was considered to be an easy condition). Trials were presented in 18 blocks of 32 trials each (6 blocks for each motor-movement condition). At the start of each block of trials, a motor-movement instruction was presented on the screen, either “Up–Down”, “Down–Up” or “Do nothing” (no-motor condition). Procedure Each trial consisted of (1) an auditory pacing signal, (2) a flashing bar with or without motor movement (i.e., the latter in the no-motor condition), and (3) the participant’s upor downward response. 1. At the start of each trial, a blue fixation dot at the left side of the screen was shown and immediately the
13
Exp Brain Res
auditory pacing signal was presented (consisting of 2 clicks with an ISI of 600 ms). The pacing signal indicated the pace at which the participants were required to press the buttons during their motor movement afterward. The fixation dot changed from blue to red 1,500 ms after the second click. 2. In the no-motor blocks, the presentation of two white horizontal bars (200-ms duration with an ISI of 600 ms) followed 400 ms after the color change of the fixation dot. In the motor upward and downward blocks, the fixation color change is the sign for the participant to start their motor movements. In the upward block, the participant pressed first the lower and then the upper button in approximately the same rhythm as the preceding auditory pacing signal (600 ms ISI). Simultaneously, with each key press, the white horizontal bar flashed for 200 ms. In the downward block, first the upper button was pressed and then the lower button. The vertical location of each bar depended on the bar-motion condition (either jumping down–up or up–down with a particular vertical displacement), while the eccentricity depended on the eccentricity condition (close or far from fixation). If the participant pressed the two buttons too fast (tapping interval 700 ms), the participant received feedback on the screen (“too fast” or “too slow”) in order to adjust their inter-tapping interval in subsequent trials. 3. After the bar flashed 2 times, the participant’s task was to judge whether the bar had moved upward or downward by using two dedicated keys with their left hand on the keyboard. The next trial started 1,000 ms after a response was detected. To acquaint participants with the task, experimental blocks were preceded by a practice session. The practice session consisted of 6 blocks (each motor-movement instruction twice) of 32 trials. At the end of the practice session, all participants were accustomed to make the motor movements in the correct rhythm.
Results Trials of the practice session were excluded from the analyses. The average interval between key presses was 591 ms, which indicates that participants performed the motor movements in the correct rhythm. For each participant, the proportion of “upward” responses was calculated for each combination of bar motion (−.525°, −.375°, −.225°, −.075°, .075°, .225°, .375°, .525° upward), eccentricity (close, far), and motor movement (upward, downward, no motor). For each combination of motor movement and
Exp Brain Res Table 1 Average PSEs and JND values for the different motor-movement conditions at close and far eccentricities (SEM between parentheses) Eccentricity
Motor movement PSE
JND
Close to fixation (easy)
Upward Downward No motor Upward
.07 (.00) .07 (.00) .08 (.00) .16 (.02)
Far from fixation (difficult)
Downward No motor
−.07 (.02) .01 (.01) .02 (.02) −.20 (.05) .11 (.03)
.17 (.02)
−.06 (.03)
.20 (.03)
The PSE represents the bar motion in degrees at which the participant perceives the bar as static. The JND represents the displacement in degrees at which 75 % of the responses is correct
eccentricity, an individually determined psychometric function was calculated by fitting a cumulative normal distribution over the eight bar-motion data points using maximum likelihood estimation. One participant was eliminated from the analysis because no reliable psychometric function could be estimated (this participant had a rigid bias toward downward responses even when the bar motion was large and clearly perceivable for other participants; N = 14 remained). The mean of the resulting distribution (the interpolated 50 % crossover point) is the point of subjective equality (henceforth PSE), which represents the bar motion in degrees at which the participant perceives the bar as static (50 % up and 50 % down responses). Besides the PSE, we calculate the just noticeable difference (henceforth JND), which is inversely related to the distribution’s slope and is a measure of a participant’s sensitivity in distinguishing visual upward and downward bar motions. The JND represents the displacement in degrees at which 75 % of the responses is correct. The average PSEs and the JNDs are shown in Table 1 (and see Fig. 1c for a graphical representation of the PSEs). A repeated measures overall ANOVA with motor movement (upward, downward, no motor) and eccentricity (close, far) as within-subject factors was performed on the PSEs and JNDs. JND. The ANOVA on the JND data revealed no main effect of motor movement, F(2,12) = 1.05, p = .37, and no significant interaction effect, F
