Stimulus-response compatibility in the approach-avoidance task interacts with go/no-go signals.
Ting Yat Wong Department of Psychology University of Amsterdam, The Netherlands
In the fulfillment of research master in psychology program Thesis Supervisor: Dr. Mark Rotteveel
Abstract
Goal-directed behaviors are affected by stimuli, motivated actions and go/no-go decisions. Former research displayed that affect congruent movement (i.e. positive-approach & negative-avoidance) can be performed faster than incongruent movement (Rotteveel & Phaf, 2004). It is also known that stop signals contains a negative connotation (Buttaccio & Hahn, 2010). However, how affect and motivated actions interact with go/no-go signals remains unclear. The present study compared go-signal response times (GSRT) and stop-signal response times (SSRT) of behavioral inhibition, which is estimated by the stop signal task, between groups received congruent instructions (CO) and incongruent instructions (IC). Results revealed that CO group participants responded faster than IC group’s in both go and stop conditions.
Running Head: Affect, Actions And Go/No-Go
Introduction
An important function of executive control is to suppress potential behaviors when they are not relevant to the current goals and intentions (i.e. task demands). Response inhibition is an example of such behavioral inhibition which can serve to constrain oneself. For instance, stopping yourself from approaching your favorite donuts when you are on a diet. Preventing from consuming high calorie food aligns the current goal while sweat food that you are craving for may generate positive feelings and an initial approach tendency toward it. This example consists of three major components: affectively valenced stimuli (donuts), motivated action tendencies (reaching the donuts) and go/no-go decisions (restraining to consume it). According to our knowledge, however, till now no study exists to examine how these three components could interact with each other and influence the subsequent final action we carry out. We hypothesized that most probably affective connotations of all three components and their match or mismatch with each other could determine ease of processing which may either facilitate or hinder the corresponding response time of the subsequent actions.
Affect supports an inherited advantage in gene survival by translating external environment conditions into internal representations that help to discriminate beneficial and detrimental conditions spontaneously (Johnston, 2003). The most apparent evidence can be observed in an affective congruent arm movement within an approach-avoidance task (AAT). When performing an affective congruent arm movement, participants are required to perform approach behaviors (i.e., arm flexion) when seeing positive valenced stimuli (PS) or avoidance behaviors (i.e., arm extension) when seeing negative valenced
stimuli (NS). In contrast, with an incongruent arm movement instruction participants should approach in response to NS or avoid in response to PS. Reaction times in arm movement are generally faster in congruent pairs than incongruent ones (e.g., Chen & Bargh, 1999; Rotteveel & Phaf, 2004) and these results have been repeatedly replicated using different paradigms, such as joystick tasks (Chen & Bargh, 1999; but for a non-replication see Rotteveel et al., in press), joystick-feedback tasks which created an illusion that images moved closer or further (Rinck & Becker, 2007), and manikin tasks (Krieglmeyer & Deutsch, 2010). In most of these paradigms, some kind of arm movement is required. Taking joystick tasks as an example, when pushing the joystick the arm extends and it is like putting undesired things away (avoidance); when pulling the arm flexes and it is like bringing desired things closer (approach; Cacioppo, Priester, & Berntson, 1993). As human beings already perform such arm flexion and extension time after time in their lifetime, it is assumed that affectively valenced stimuli can automatically activate arm flexion and arm extension (Chen & Bargh, 1999; for a review, see Krieglmeyer, De Houwer, & Deutsch, 2013; Krieglmeyer, Deutsch, De Houwer, & De Raedt, 2010). Although a joystick AAT uncovered evidence that support affect-congruent arm movement effect, unwanted movement whenever pushing or pulling the joystick may somewhat bias the results. Rotteveel and Phaf (2004) invented a three-button vertical stand device which it can avoid unwanted arm movement to test how affectively valenced stimuli influence the response latencies of arm flexion and arm extension. In this setup, only biceps and triceps muscles are involved in performing approach (pressing upper buttons) and avoidance (pressing lower buttons) behaviors. Furthermore, biceps and triceps are stretched equally before performing any responses when holding the middle (i.e. home) button. Affect-congruent movements were
also faster here (experiment 1): arm flexion was faster with happy faces than with angry faces whereas arm extension was faster with angry faces with happy faces.
Affect is not only evoked by valenced stimuli per se but can also result from the mere monitoring processes of internal functioning and information processing, according to the affective monitoring hypothesis (Phaf & Rotteveel, 2009, 2012, see also Carr, Rotteveel & Winkielman, submitted). Match, familiarity, and congruence can produce positive affect; in contrast, mismatch, novelty, and incongruence can generate negative affect (Phaf & Rotteveel, 2009). Affect induced by the monitoring process can influence subsequent emotional ratings and, for instance, behavioral response times such as in the AAT. For example, left to right readers (e.g. Dutch participants) rated arrows pointing to the right as more positive than arrows pointing to the left (Phaf & Rotteveel, 2009). The match of reading direction and arrows generates positive affect; in contrast, mismatch of them induce negative affect. In the same study, the researchers further demonstrated that induced affect can be reflected in AAT performance as avoidance latencies. These results suggested that congruence between reading direction and the arrow direction induced affect that not only impacted explicit affective evaluation (Experiment 1) but even approach and avoidance behaviors, or more specifically arm flexion and extension, in the absence of any explicit affective evaluation (Experiment 2). Based on this idea, we expected if the affective connotations of the aforementioned components match each other, i.e. all positive or all negative connotations, a positive affect will be generated, vice versa. Thus, a quicker response time of subsequent movement would be observed in the congruent conditions.
In real-life situations, behavioral responses toward emotional stimuli may not always be appropriate in every context. An inhibition system is essential to guide our behaviors in an adequate fashion in everyday life. One way to study inhibition and its influences can be
through employing a stop-signal task (SST; Logan, Cowan, & Davis, 1984). This task requires participants to perform an action according to a given instruction, for instance, pressing a button if they see happy faces (i.e., go conditions). Meanwhile, whenever receiving a cue, which can be either auditory or visual, participants should withhold their ongoing actions (i.e., stop conditions). Unlike go-signal response times (GSRT), stop-signal response times (SSRT) cannot be measured directly. However, based on the horse-race model proposed by Logan (1984), stop-signal response times can be estimated (see Verbruggen & Logan, 2008 for a more detailed review).
Previous studies found that withholding a response cued by a stop signal may lead to devaluation of the stimuli (Buttaccio & Hahn, 2010; Doallo et al., 2012; Fenske, Raymond, Kessler, Westoby, & Tipper, 2005; Kiss, Raymond, Westoby, Nobre, & Eimer, 2008; Veling, Holland, & van Knippenberg, 2008; Wilkowski & Robinson, 2006). For example, Buttaccio and Hahn (2010) conducted three experiments to verify how behavioral inhibition influenced the later affective evaluation of novel shapes (neutral stimuli). Results revealed that novel shapes following inhibition were rated as more unpleasant than those following without inhibition. Such experiments suggested that stop behavior has negative affective consequences. Veling and colleagues (2008) also found similar results that withholding responses may lead to subsequent devaluation of stimuli. In their experiment, however, this devaluation effect was found only for positive stimuli but not for neutral or negative stimuli. These inconsistent results may be explained by affective monitoring when processing affective connotations of the aforementioned components. Within similar stop signal studies, participants were instructed to press the button if no stop-signal was presented. In this context, either an approach or avoidance behaviors has been done is vague to the participants. The implicit psychologically meaning of pressing the button could be “approach
the target stimulus”. Since the approach tendency may contain positive affective connotations, this may not match with the stop signal that contain negative affective connotations. Based on affective monitoring notions, negative affect will then be generated and lead to negative evaluation of the stimuli at hand. If the stimulus is positive, such incongruence may be extended further, resulting in the findings that stopped signals only influence positive stimuli. This missing components may explain why an inconsistent result is documented. In our research we want to explore this explanation further by using two different movements (i.e., approach and avoidance) that can be performed or stopped in one experimental design.
In the present study, the main research question concerns the interplay between affectively valenced stimuli, action tendencies (and movement) and go/no-go signals based on their affective connotations. Specifically, we asked the question how GSRT and SSRT can be modulated by the interaction of these components mapped on two different behavioral affectively connotated responses (i.e., flexion and extension). According to the affective monitoring hypothesis, the more congruent components in terms of affective connotations are, the easier the processing is. We argued that stop signals not only hamper an approach response (arm flexion) with positive stimuli but can also facilitate a specific avoidance response (arm extension) with negative stimuli. If so, facilitation should be observed not only in positively congruent components (i.e., positive-flexion-go) but also in negatively but congruent components (i.e., negative-extension-stop). Therefore, we predicted that, first of all, in a go-condition, shortest latencies would be observed when all three components match with each other (i.e., positive-flexion-go) when compared with other incongruent pairs (i.e., positive-extension-go, negative-flexion-go and negative-extension-go). Secondly, in a stop-condition, shorter latencies would be observed when all three components match
with each other (i.e., negative-extension-stop) when compared with other incongruent pairs (i.e., negative-flexion-stop, positive-flexion-stop and positive-extension-stop).
Method
Participants
Total 35 participants (9 males) were recruited in the experiment. All recruited participants had normal or corrected-to-normal sight and no hearing problem and right handed. The mean age of the male participants was 26.3 (SD = 4.56) and that of the female participants was 22.8 (SD = 3.02). Participants whose correct response rate of categorization within a condition exceeds ±2.5 standard deviation of the mean would be considered as outliers and excluded from the analysis. Moreover, participants whose response times exceeded ±2.5 standard deviation of the mean within each condition will be excluded as well. As a result, five participants had been excluded from the analysis because of low correct response rates in either go or stop conditions.
Procedure
Participants were first welcomed and briefed once they arrived at the laboratory. Instructions about the experimental procedure were then given and written informed consent was obtained. Participants were invited to sit in front of a computer (eye-to-monitor distance: ~60 cm) and wear a headphone.
Participants were randomly and evenly assigned into 2 groups. Group A were given a congruent instruction which they were instructed to press the upper button (flexion) if they saw a happy face and press the lower button (extension) if they saw an angry face. Group B received an incongruent instruction which if a happy face was presented, participants were
instructed to press the lower button (extension) and if an angry face was presented, they were instructed to press the upper button (flexion). The button to button distance was fixed to 8.7cm. A total of 50 (25 males and 25 females) happy and angry faces (headshots and frontal view) were selected from Karolinska set (Karolinska Directed Emotional Faces; Lundqvist, Flykt, & Öhman, 1998) based on their correct response rates and intensity from a recent validation study (Goeleven, De Raedt, Leyman & Verschuere, 2008).
The AAT device by Rotteveel and Phaf (2004) was adjusted to fit each participant individually. Participants were instructed to always hold the middle button at the beginning of each trial and release it when they responded to the upper and lower button. When holding the middle button, two major muscles of the arm, triceps and biceps, were equally stretched. When the participants pressed the upper button, the biceps contracts and arm flexion was the result (i.e., approach); when the participant pressed the lower button, the triceps contracted, resulting in arm extension (i.e., avoidance). Using this experimental setup enabled us to disentangle initiation time from movement time. Initiation time (IT) is the latency from when the stimulus is presented to leaving the middle button and reflects action preparation mirroring actual action tendencies. Movement time (MT) measures the latency between leaving the middle button and pressing the correct response button mirroring execution of action itself. The RTtotal was the sum of IT and MT.
Twelve practice trials (4 stop trials) were given on a 20 inch monitor (resolution: 1680 x 1050 pixels) to the participants and they were required to give responses according to the above instructions. Feedback were given to the participants, indicating their performance and the correct answers if applicable. After that, an experimental block with 200 trials followed. The experimental block was basically identical to the practice trials except that there were no feedback. Each trial started with a white fixation cross presented on a black
background (RGB: 0, 0, 0) for 500ms. Then either a happy or angry face (size: 400 x 600 pixels) appeared at the center of the screen for 1000ms, resulting in a visual angle of 17° x 23.8°. Participants were required to press the corresponding button as quickly and correctly as possible. On a one-fourth of the trials, an auditory tone binaurally over headphones appeared after the presentation of the faces. This tone, presented at 1000 Hz with an intensity of 60dB for 100ms (rise and fall time 16ms), acted as the stop signal, prompting participants to inhibit their response. If the auditory cue was heard, participants should attempt to withhold that response and not press either the upper or lower button. Stop signals occurred equally for each emotion; therefore, a total of 50 trials (25 for each facial expression) were accompanied with the stop signal in each group. In addition, participants were told to treat no signal as a go signal in order to make go signals more explicitly. Figure 1 displayed a graphical illustration of the experiment.
[Insert Figure 1 Here]
The delay between the presentation of the stimulus and the auditory stop signal was varied trial-by-trial using a staircase tracking method which either increased or decreased the stop signal delay (SSD) by 50ms for the next stop signal trial depending on whether the participant successfully inhibited (delay increased) or failed to inhibit (delay decreased) their response to the stimulus. The SSD was set to be 250ms at the start of the experiment. The tracking method was to converge on a SSD where participants successfully inhibit on ∼50% of stop-signal trials (see Logan, Schachar, & Tannock, 1997 and Verbruggen & Logan, 2008 for a detailed description). SSRT was estimated by subtracting the mean SSD latency from the mean GSRT that was calculated from correct go-signal (without the tone) trials (i.e. SSRT = mean GSRT - mean SSD).
To prevent participants from waiting for the tone in order to have a better performance, they were told that the analysis would focus on the response times of go-trials rather than the successful rate of withholding their responses (Verbruggen & De Houwer, 2007).
Repeatedly doing a lot of trials using the stand device we used in this study caused muscle fatigue and pain. To reach the minimum stop-signal trials for estimating the SSRT, therefore, participants were invited to perform the task again a week later which the procedure was exactly the same as the first session. In total, participants performed 300 go-trials (150 for each facial expression) and 100 stop-go-trials (50 for each facial expression).
Results
Correct Response Rates
In a go condition, the correct response rate of participants in the congruent group was 97.6% (SD = .027) and that of participants in the incongruent group was 96.3% (SD = .043). In a stop condition, the successful rate of withholding their responses in the congruent group was 58.2% (SD = .054) and that in the incongruent group was 58.4% (SD = .093). A between-group comparison showed that none of groups performed better than each other (all ps > .05).
Release Times (RT)
A 2 x 2 repeated measures mixed ANOVA with Emotion (happy and angry) as a within-subject factor and Group (congruent and incongruent instructions) as a between-within-subject factor was performed on RT, MT, SSRT and RTtotal respectively. A main effect of emotional
expressions on RT was found (F1,28 = 6.22, p = .019, 𝜂2 = .182). Faster RT was observed with
Comment [TW1]:
Mark:
Of course you can report this post hoc analysis but I think the most interesting difference that should be tested is flexion with happy and angry expressions as well extension with happy and angry indicative of congruency mapped on both arm movements. Also because a similar comparison can be made with the stop trials: in both cases congruent trials are faster and easier to stop
TY Replied:
The mapping of different movement and different facial expressions is fixed due to the experiment setting. Congruent conditions (flexion with happy faces and extension with angry faces) and incongruent conditions (flexion with angry faces and extension with happy faces) were given to two different groups of participants. Therefore, the comparison are done like this fashion.
happy facial expressions. Between-group comparison showed that participants who received congruent instructions responded faster than those who received incongruent instructions (F1,28 = 8.37, p = .007, 𝜂2 = .23). An interaction effect between expressions and
groups was observed (F1,28 = 22.5, p < .001, 𝜂2 = .445). The post hoc analysis further
indicated that participants who received congruent instructions responded faster in happy than angry expressions (t = 4.15, p = .001) while those who received incongruent instructions responded marginally faster in angry than happy expressions (t = -1.98, p = .065; see Figure 2).
[Insert Figure 2 here]
Movement Times (MT)
An emotion main effect was not found on MT (F1,28 = .418, p = .523, 𝜂𝑝2 = .015) but a
significant between-group effect (F1,28 = 9.07, p = .005, 𝜂𝑝2 = .245) and interaction effect
between groups and emotions (F1,28 = 9.39, p = .005, 𝜂𝑝2 = .251) were observed. Faster MT
was observed in participants who received incongruent instruction than those who received congruent one (see Figure 3). Post hoc analysis further indicated that in a congruent condition, no MT difference was observed between happy and angry expressions (t = -1.49,
p = .161). However, in an incongruent condition group, participants responded faster to
happy than angry expressions (t = 3.002, p = .008). [Insert Figure 3 here]
Stop Signal Response Time estimated by Release Times (SSRT)
There was no emotion main effect on SSRT (F1,28 = 1.15, p = .293, 𝜂𝑝2 = .039) but a
between-group effect was documented (F1,28 = 5.91, p = .022, 𝜂𝑝2 = .174). Participants who
received a congruent instruction had a faster SSRT than those who received an incongruent one. An interaction effect between group and emotion was only marginally significant (F1,28
= 3.43, p = .075, 𝜂𝑝2 = .109). Post hoc analysis indicated that SSRT did not differ between
happy and angry expressions in two groups (t = 1.77, p = .102; t = -.644, p = .529). Figure 4 illustrated the SSRT results.
[Insert Figure 4 here]
Total Response Times (RTtotal)
Figure 5 illustrated the RTtotal results. The emotion main effect on RTtotal was only
marginal (F1,28 = 3.58, p = .069, 𝜂𝑝2 = .113). No significant between group effect and
emotion-group interaction effect was found (F1,28 = 2.39, p = .134, 𝜂𝑝2 = .078; F1,28 = .756, p
= .392, 𝜂𝑝2 = .026).
[Insert Figure 5 here]
Stop-Signal Response Time estimated by Total Response Time (SSRTtotal)
No significant emotion main effect on SSRTtotal was only marginal (F1,28 = 2.51, p = .124,
𝜂𝑝2 = .082). Between-group effect and emotion-group interaction effect was also
insignificant (F1,28 =.613, p = .440, 𝜂2 = .021; F1,28 < .001, p = .995, 𝜂𝑝2 < .001). Figure 6
illustrated the SSRTtotal results.
[Insert Figure 6 here]
Discussion
The present study addressed the question whether the interaction between go/no-go stop signals and affect congruent/incongruent movement (stimulus-response, S-R, compatibility) exists and influences the subsequent behavioral response time. The answer is positive: relatively faster response times are observed in affect congruent movement (i.e., positive-approach & negative-avoidance) in both the go and no-go signals.
Go-trials results
Congruency effect is commonly observed in various tasks, such as Simon task (e.g., Simon & Rudell, 1967), flanker task (e.g., Eriksen & Eriksen, 1974) and Stroop task (e.g., Stroop, 1935) and AAT (e.g., Chen & Bargh, 1999; Rotteveel & Phaf, 2004). Faster response times has been documented in congruent conditions. For example, in Simon task, response times are faster when arrows and response locations matched each other (e.g., right arrow & right key). According to affect monitoring hypothesis, positive affect is generated by the ease of processing. Such compatible stimulus and response mappings in, for example, a Simon task result in a faster response time.
This congruency effect was also documented in the present study. In line with our prediction, in go-trials, faster RT were recorded in the congruent trials than in the incongruent ones. This result is consistent to previous studies (e.g., Phaf & Rotteveel, 2009; Rotteveel & Phaf, 2004). One advantage of the present study is that we could isolate initiation times from the total response times RTtotal. This congruency effect can only be
observed in the prepotent action (leaving the middle button) but not the complete action (from leaving the middle button to either the upper or lower bottom). This suggested that the congruency effect happened in an early stage (preparation stage) of our responded actions. As suggested by Johnston (2003), affect helps to discriminate beneficial and detrimental conditions for a survival purpose. Therefore, affective influence should impact more on the bottom-up processing rather than the top-down processing.
It is noteworthy that participants responded even faster in happy-flexion-go trials than in angry-extension-go trials. This is in favor of our hypothesis that congruent affective connotations of stimulus, response and signals are the easiest processing condition among the others and lead to relatively faster RT.
In previous studies, we did not discover any difference in participants’ MT in a normal AAT (e.g., Rotteveel & Phaf, 2004); however, we observed a significant difference of MT between two groups in the present study. Faster MT were observed in the incongruent group than in the congruent group. Although participants needed longer preparation times for incongruent S-R mappings in order to execute corresponding movement, a faster movement time was observed. This suggests that initial longer preparation due to incongruent information does not necessarily sacrifice the overall performance. Thus, we did not find any difference on the RTtotal between two groups. This result may also suggest
differentiated influence from congruency and conflict of the S-R mappings. When affect congruent movement is performed, positive affect which leads to RT is generated and the process of response selection is after initiation of the response which leads to slower MT (bottom-up influences). However, when affect incongruent movement is performed, conflict which leads to slower RT arises and selection are made during the initiation which leads to a faster MT (top-down influences).
No-go trials results
The SSRT result did not match our hypothesis but it demonstrated that stop signal did not interact with the stimuli and responses in terms of affective connotations. Instead, the interaction was found between the signal and the S-R compatibility.
To resolve conflict in congruency tasks, a main area in the prefrontal cortex – anterior cingulate cortex (ACC) is involved (see Etkin, et.al, 2006, Carter & van Veen, 2007 for a review). Anterior cingulate cortex (ACC) activation has been documented in different forms of congruency tasks, such as a flanker task (e.g., Botvinick, et al., 1999), a Simon task (e.g., Peterson, et al, 2002), and this activation has been also observed in a go/no-go task (e.g., Durston, 2002). In AAT, Roelofs and colleagues (2009) recorded increased activity in left
lateral orbitofrontal cortex (OFC) when participants performed affect-incongruent movement. The activated region involved in congruency tasks is different from that in the AAT. This implies the processing of conflict in this two types of tasks may be different. Unlike the aforementioned congruency tasks, the AAT does not contain any irrelevant information. More importantly, it is more about the direct motivational association between stimuli and responses. In AAT, the stimulus valence interferes with the arm flexion and extension movement directly, facilitating or slowing subsequent behavioral responses. Thus, affect congruent/incongruent movement is more influenced by the motivational system. This explains why different brain activation areas may be involved.
Using a franker task or other types of congruency tasks, faster SSRT have been observed in the congruent conditions (e.g., Verbruggen, Liefooghe, & Vandierendonck, 2004). The researchers interpreted this effect indicated that inference control and response inhibition may rely on the same monitoring mechanisms, namely the conflict monitoring system (Botvinick, et al., 2001). We also found a similar effect in affect congruent movement under no-go signals, suggesting resolving conflicts in affect incongruent movement may require a similar mechanism as in, for instance, Simon tasks, flanker task or Stroop task which this will also interact with the stop signal.
When incongruent S-R mappings are given to the participants, conflict arises. When adding another conflict, a stop signal in our study, two types of conflict arise simultaneously. Processing two types of conflict may result in a slower response time. Unlike go-signal, no-go signal or stop signal per se will activate conflict monitoring system. This is the fundamental difference between two conditions. This may the reason why we only find the congruency effect in the go-trials.
The current study suggested that the congruency effect can be observed in the interactions between S-R mappings and go signals. However, this congruency effect is more based on the stimulus or responses that activate a bottom-up processing. Stop signals, on the other hand, may require top-down processing to resolve the conflict. If S-R mappings are incongruent, conflict arise. This may explain why compatible S-R mappings also resulted in a faster response time in no-go trials.
Limitation
Few participants reported in the exit interview that they had their left hand (non-responded hand) to support the right hand occasionally because of feeling fatigue and pain in the second half of the experimental trials. These might affect the estimation of SSRT. Due to the experiment set-up limitation, we did not map flexion and extension movements with happy and angry faces within a participants. We therefore could not directly compare the difference between their response time.
Conclusion
To conclude, affect congruent responses are faster in go as well as in no-go trials. The present study revealed that congruency effect of compatible S-R mappings is observed not only in go trials but also in stop trials, and the interaction effect between S-R mappings is only observed in the go signals but not in stop signals, indicating there is a difference in processing when two types of conflict arise simultaneously.
References (not yet finalized)
Botvinick, M. M., Bravers, T. S., Barch, D. M., Carter, C. S., & Cohen, J.D. (2001). Conflict monitoring and cognitive control. Psychological Review, 108(30), 624-652.
Botvinick, M., Nystrom, L. E., Fissell, K., Carter, C. S., & Cohen, J.D. (1999). Conflict monitoring versus selection-for-action in anterior cingulate cortex. Nature, 402, 179– 181. doi:10.1038/46035
Buttaccio, D. R., & Hahn, S. (2010). The effect of behavioral response on affective evaluation. Acta Psychologica, 135(3), 343–348. doi:10.1016/j.actpsy.2010.09.004 Cacioppo, J. T., Priester, J. R., & Berntson, G. G. (1993). Rudimentary determinants of
attitudes: II. Arm flexion and extension have differential effects on attitudes. Journal of
Personality and Social Psychology, 65(1), 5–17. doi:10.1037/0022-3514.65.1.5
Carter, C. S., & van Veen, V., (2007). Anterior cingulate cortex and conflict detection: An update of theory and data. Cognitive, Affective, & Behavioral Neuroscience, 7(4), 367-379.
Chen, M., & Bargh, J. (1999). Consequences of automatic evaluation: Immediate behavioral predispositions to approach or avoid the stimulus. Personality and Social Psychology
Bulletin, 25(2), 215–224. doi:10.1177/0146167299025002007
Chiu, Y. -C., Cools, R., & Aron, A. R. (2014). Opposing effects of appetitive and aversive cues on go/no-go behavior and motor excitability. Journal of Cognitive Neuroscience, 26(8), 1851-1860. doi:10.1162/jocn_a_00585
Doallo, S., Raymond, J. E., Shapiro, K. L., Kiss, M., Eimer, M., & Nobre, A. C. (2012). Response inhibition results in the emotional devaluation of faces: neural correlates as revealed by fMRI. Social Cognitive and Affective Neuroscience, 7(6), 649–659. doi:10.1093/scan/nsr031
Durston, S. Thomas, K. M., Worden, M. S., Yang, Y., & Case, B. J. (2002) The effect of preceding context on inhibition: An event-related fMRI study. NeuroImage, 16, 449– 453. doi:10.1006/nimg.2002.1074
Etkin, A., Egner, T., Peraza, D. M., Kandel, E., R., & Hirsch, J. (2006). Resolving emotional conflict: A role for the rostral anterior cingulate cortex in modulating activity in the amygdala. Neuron, 51(7), 871-882. doi: 10.106/j.neuron.2006.07.029
Eriksen, B. A., & Eriksen, C. W. (1974). Effects of noise letters upon identification of a target letter in a nonsearch task. Perception and Psychophysics, 16, 143–149.
Fenske, M. J., Raymond, J. E., Kessler, K., Westoby, N., & Tipper, S. P. (2005). Attentional inhibition has social-emotional consequences for unfamiliar faces. Psychological
Science, 16(10), 753–758. doi:10.1111/j.1467-9280.2005.01609.x
Goeleven, E., De Raedt, R., Leyman, L., & Verschuere, B. (2008). The Karolinska Directed Emotional Faces: A validation study. Cognition & Emotion, 22(6), 1094-1118. doi: 10.1080/02699930701626582
Johnston, V. (2003). The origin and function of pleasure. Cognition & Emotion, 17(2), 167– 179. doi:10.1080/02699930302290
Kiss, M., Raymond, J. E., Westoby, N., Nobre, A. C., & Eimer, M. (2008). Response inhibition is linked to emotional devaluation: Behavioural and electrophysiological evidence.
Frontiers in Human Neuroscience, 2(October), 1–9. doi:10.3389/neuro.09.013.2008
Krieglmeyer, R., De Houwer, J., & Deutsch, R. (2013). On the nature of automatically triggered apporach-avoidance behaviors. Emotion Review, 5(3), 280–284. doi:10.1177/1754073913477501
Krieglmeyer, R., & Deutsch, R. (2010). Comparing measures of approach-avoidance behaviour: The manikin task vs. two versions of the joystick task. Cognition & Emotion,
24(5), 810–828. doi:10.1080/02699930903047298
Krieglmeyer, R., Deutsch, R., De Houwer, J., & De Raedt, R. (2010). Being moved: Valence activates avoidance behavior independently of evaluation and approach-avoidance intentions. Psychological Science, 21(4), 607–613. doi:10.1177/0956797610365131
Logan, G. D., Cowan, W. B., & Davis, K. a. (1984). On the ability to inhibit simple and choice reaction time responses: a model and a method. Journal of Experimental Psychology.
Human Perception and Performance, 10(2), 276–291. doi:10.1037/0096-1523.10.2.276
Logan, G. D., Schachar, R. J., & Tannock, R. (1997). Impulsivity and inhibitory control.
Psychological Science, 8(1), 60–64. doi:10.1111/j.1467-9280.1997.tb00545.x
Lundqvist, D., Flykt, A., & Öhman, A. (1998). The Karolinska Directed Emotional Faces - KDEF, CD ROM from Department of Clinical Neuroscience, Psychology section, Karolinska Institutet, ISBN 91-630-7164-9.
Peterson, B. S., Kane, M. J., Alexander, G. M., Lacadie, C., Skudlarski, P., Leung, H. C., … , Gore, J. C. (2002). An event-related functional MRI study comparing interference effects in the Simon and Stroop tasks. Cognitive Brain Research, 13(3), 427-440. doi: 10.1016/S0926-6410(02)00054-X
Phaf, R. H., & Rotteveel, M. (2009). Looking at the bright side: The affective monitoring of direction. Emotion, 9(5), 729–733. doi:10.1037/a0016308
Phaf, R. H., & Rotteveel, M. (2012). Affective monitoring: A generic mechanism for affect elicitation. Frontiers in Psychology, 3(March), 47. doi:10.3389/fpsyg.2012.00047
Rinck, M., & Becker, E. S. (2007). Approach and avoidance in fear of spiders. Journal of
Behavior Therapy and Experimental Psychiatry, 38(2), 105–120.
doi:10.1016/j.jbtep.2006.10.001
Roelofs, K., Minelli, A., Mars, R. B., van Peer, J., Toni, I. (2009). On the neural control of social emotional behavior. SCAN, 4, 50-58.
Rotteveel, M., & Phaf, R. H. (2004). Automatic affective evaluation does not automatically predispose for arm flexion and extension. Emotion, 4(2), 156–172. doi:10.1037/1528-3542.4.2.156
Rotteveel, M., Gierholz, A., Pinto, Y., Matzke, D., Streingroever, H., Verhagen, J., … Wagenmakers, E.-J. (in press). On the automatic link between affect and tendencies to approach and avoid: Chen and Bargh (1999) revisited. Frontiers in Psychology.
Simon, J. R., & Rudell, A. P. (1967). Auditory S-R compatibility: The effect of an irrelevant cue on information processing. Journal of Applied Psychology, 51, 300-304.
Stroop, J. R. (1935). Studies of interference in serial verbal reactions. Journal of Experimental
Psychology, 18, 643–662.
Veling, H., Holland, R. W., & van Knippenberg, A. (2008). When approach motivation and behavioral inhibition collide: Behavior regulation through stimulus devaluation. Journal
of Experimental Social Psychology, 44(4), 1013–1019. doi:10.1016/j.jesp.2008.03.004
Verbruggen, F., Liefooghe, B., Vandierendonck, A. (2004) The interaction between stop signal and distractor interference in the flanker and Stroop task. Acta Psychologica, 116(1), 21-37. doi: 10.1016/j.actpsy.2003.12.011
Verbruggen, F., & De Houwer, J. (2007). Do emotional stimuli interfere with response inhibition? Evidence from the stop signal paradigm. Cognition & Emotion, 21(2), 391– 403. doi:10.1080/02699930600625081
Verbrugen, F., Liefooghe, B., Notebaert, W., & Vandierendonck, A. (2005). Effects of stimulus-stimulus compatibility and stimulus-response compatibility on response inhibition. Acta Psychologica, 120(3), 307-326. doi: 10.1016/j.actpsy.2005.05.003
Verbruggen, F., & Logan, G. D. (2008). Response inhibition in the stop-signal paradigm.
Trends in Cognitive Sciences, 12(11), 418–424. doi:10.1016/j.tics.2008.07.005
Wilkowski, B. M., & Robinson, M. D. (2006). Stopping dead in one’s tracks: Motor inhibition following incidental evaluations. Journal of Experimental Social Psychology, 42(4), 479– 490. doi:10.1016/j.jesp.2005.08.007
Winkielman, P., & Cacioppo, J. T. (2001). Mind at ease puts a smile on the face: Psychophysiological evidence that processing facilitation elicits positive affect. Journal
of Personality and Social Psychology, 81(6), 989–1000.
Appendix
Figure 1. Experimental Setup: An example of happy facial expression.
*Staircase method: The stop delay (SD) changed depending on the performance of participants. If they successfully withhold their response in a previous stop trial, SD will be increased by 50ms so the stop task becomes more difficult. In contrast, if they fail to withhold their response, SD will be decreased by 50ms (easier).
Figure 2. Release times (RT)
*Significant difference at .05 level + Significant difference at .01 level
Figure 3. Movement times (MT)
*Significant difference at .05 level
Figure 4. Stop-signal response times (SSRT)
*Significant difference at .05 level
Figure 5. Total response times (RTtotal)
*Significant difference at .05 level
Figure 6. Stop signal response times using total response to estimate (SSRTtotal)
Note to myself
Our current results are partly consistent with the result that appetitive cues speed up the motor system and promote action while aversive cues dampen the motor system and promote inaction (Chiu, Cools, & Aron, 2014). In our result, we found that participants also stop quicker in compatible S-R mappings which is more positive than incompatible S-R ones.
Some Simon task studies did not find an effect unless considering trial sequence into account (e.g. Verbrugen, Liefooghe, Notebaert & Vandierendonck, 2005). In contrast with these results, we find a congruency effect in stop trials without analyzing the trial sequence. This difference may be due to establish the association between stimulus
