https://doi.org/10.1177/1747021818807181

Quarterly Journal of Experimental Psychology 1 –15© Experimental Psychology Society 2018Article reuse guidelines: sagepub.com/journals-permissionsDOI: 10.1177/1747021818807181qjep.sagepub.com

Introduction

According to ideomotor theories, the acquisition of volun-tary action control happens in two stages (Elsner & Hommel, 2001). These two stages are the basis of the ideo-motor mechanism (Elsner & Hommel, 2001; Greenwald, 1970; Harless, 1861; Herbart, 1825; Hommel, 2009; Hommel, Müsseler, Aschersleben, & Prinz, 2001; James, 1890; Lotze, 1852). The first stage is conceived as a phase of integration involved by an ideomotor bidirectional asso-ciation. Within this phase, sensory effects resulting from random movements automatically come to be distin-guished as specific consequences of this movement. If the same sensory effect follows the same movement repeat-edly enough, movement and effect become integrated

together in a common structure. Integration is understood in this framework as the result of the ideomotor bidirec-tional association, by which motor and perceptive events are linked. This bidirectional association becomes appar-ent during the second phase, where the activation of one event (movement or effect) automatically induces the

How movement direction shapes the spatial representation of its effects: About the consequence of the ideomotor bidirectional association

Hélène Lestage1, Thomas Camus2, Vincent Dru1 and Thibaut Brouillet1

AbstractIdeomotor theories assume that action and perception share a common representational system in which a movement and its effect are equally represented and integrated by a bidirectional association. However, there is no mention of how this association leads to influence the representational content of each part. In this article, we investigated the influence of movement properties on the spatial representation of auditory effects. In line with the Action Constrains Theory of space perception, we suggest that changes in the movement direction leads to correlative changes in the spatial representation of the effect. In a pre-experiment, we replicated traditional ideomotor results with a response-effect (R-E) compatibility procedure. In two experiments, we used one condition of this procedure (i.e., the corresponding R-E mapping) to manipulate the movement properties associated to a non-spatialised effect. In the first experiment, the effect was associated with horizontal outward movements or with forward–backward movements. In the second experiment, we tested some alternative explanations for the results obtained in the first experiment. Globally, we showed that rightward movements led to localised auditory effect more on the right space than leftward movements and that backward movements led to localisation of the effect closer from the subjects than forward movements. In accordance with the Action Constrains Theory of space perception, these data suggest that movement shapes the spatial organisation of the effect representation.

KeywordsIdeomotor mechanism; action effect; bidirectional association; movement; integration; spatial representation

Received: 8 March 2018; revised: 19 September 2018; accepted: 24 September 2018

1 CeRSM Laboratory (EA2931), Paris-Nanterre University, Nanterre, France

2 EPSYLON Laboratory (EA 4556), Paul Valéry University, Montpellier, France

Corresponding author:Hélène Lestage, CeRSM Laboratory (EA2931), Paris-Nanterre University, 200 avenue de la République, 92000 Nanterre.Email: helene.lestage@gmail.com

10.1177_1747021818807181QJP0010.1177/1747021818807181The Quarterly Journal of Experimental PsychologyLestage et al.research-article2018

Original Article

2 Quarterly Journal of Experimental Psychology 00(0)

activation of the other. This second stage then refers to the selection of voluntary movement. As movement and effect are linked together, voluntary movement is now selected by activating the representation of its expected effect (i.e., the idea of this effect). This is the main assumption of ide-omotor theories. This assumption has been partially exper-imented by Elsner and Hommel (2001). In a response-effect (R-E) acquisition procedure, participants were required to press a right or a left key as a motor response to a visual stimulus. They experienced the co-occurrence of their responses with high-pitched or low-pitched tones (i.e., auditory effect of the response). Importantly, the tones were irrelevant to the task. In a subsequent test phase, par-ticipants were presented with both tones, this time as stim-uli. They were asked to press a left or a right button in response to the low or high frequency of the tone. For Elsner and Hommel (2001), this experimental procedure illustrates the ideomotor two-stage phases. The authors reported that responses of participants were faster when the tone had previously been the effect associated to this response during the acquisition phase. These results sug-gest that the perception of an effect (e.g., high or low pitch) activates the corresponding response (e.g., finger move-ments induced by the right or left key presses) which has been associated to it through the acquisition phase.

As Kunde (2001, p. 387) noticed, an issue raised by Elsner and Hommel (2001)’s experiment is that it does not prove that the responses performed during the second phase are selected according to the anticipation of the sen-sory effects. Indeed, as long as a stimulus precedes the response (as it is the case in both phases of Elsner and Hommel’s experiment), it is not possible to rule out the possibility that the highlighted bidirectional association actually concerns the stimulus and the response effect—versus, as it is stated by ideomotor theories, the response and its effect. To address this ambiguity, Kunde (2001) tested the influence of a R-E compatibility procedure on response selection without the presence of a stimulus. He compared corresponding versus non-corresponding R-E spatial mappings. Subjects were asked to press one of four horizontally aligned response keys. In the first case (cor-responding mapping), the location of subjects’ responses and the location of the visual effects triggered by these responses were the same, as a lamp was lighting up at the same location subjects pressed a response key. In the sec-ond case (non-corresponding mapping), the location of subjects’ responses and the location of the visual effects triggered by these responses were different, as a lamp was lighting up in an opposite location from where the response key was. Results revealed that although the effects were displayed after the responses, after a few trials correspond-ing mapping triggered faster and more accurate responses than non-corresponding mappings. Thereby Kunde’s experiment allows concluding that a response can truly be selected by the anticipation of its effects. During the last

two decades, ideomotor assumptions have been exten-sively and successfully tested by numerous studies (for a review read Shin, Proctor, & Capaldi, 2010).

Nowadays, the most popular version of ideomotor theories is the Theory of Event Coding (TEC) developed by Hommel et al. (2001) and Hommel (2009). In line with James (1890) and other ideomotor theories, TEC supports the idea that movements are represented accord-ing to their sensory effects. In addition, TEC claims that movement and effect are integrated and linked together in a common representation by an ideomotor two-stage procedure. This common representation is called an “event file.” It represents the stimulus, the stimulus con-text, the response, and the effect. Event files are parts of an individual’s sensorimotor experiences with his envi-ronment. In this way, perceptual and motor events share a common representational system by which the activa-tion of one event code can automatically trigger the acti-vation of the other event code.

A common representational system for perception and action

In line with Gibson’s (1966) ecological theory, TEC alleges that action and perception share a common sub-strate as well as a bidirectional influence between them. Perceived events (perceptions of stimuli and of action effects) and to-be-produced events (actions) are equally represented in a way that the activation of a specific motor code can be predicted by the activation of a specific cor-responding perceptive code, and reciprocally.

As Shin et al. (2010 p. 965) point out, if action and per-ception share a common representational system and exhibit a bidirectional association, then ideomotor theories should take interest in investigating how action properties impact perceptive content. A limited number of studies took interest in this last aspect (see Craighero, Fadiga, Rizzolatti, & Umiltà, 1999; Deubel & Schneider, 1996; Fagioli, Hommel, & Schubotz, 2007; Hommel & Müsseler, 2006; Hommel et al., 2001; Kirsch & Kunde, 2014; Müsseler & Hommel, 1997; Rizzolatti, Riggio, & Sheliga, 1994; Wykowska, Schubö, & Hommel, 2009). Nevertheless, in all instances, perception is only defined as perception of stimuli preceding and triggering the action, without any investigation of the bidirectional association consequences on the effect representation. For example, Kirsch and Kunde (2014) showed that the spatial localisa-tion of a stimulus could be influenced by the direction of planned movements associated to this stimulus. Similarly, Fagioli et al. (2007) showed that planning for an action leads to priming particular stimulus features codes. This is the case even when the action is not performed but only visualised (Fagioli, Ferlazzo, & Hommel, 2007). For example, seeing a grasping action prepares for the process-ing of objects involving grasping.

Lestage et al. 3

Nevertheless, those studies only show that association between a stimulus, a response and an effect can influence the perception of the stimulus. There is no mention what-soever of how the bidirectional association might affect the effect representation. This is what we intend to investi-gate in this study.

The correlative influence of movement properties on the effect representation: our hypothesis

Action Constraints Theories of space perception (ACT, for a review see Morgado & Palluel-Germain, 2015) sug-gest that the properties of our intended actions modify the spatial representation of the stimuli we mean to identify. In other words, for ACT, the way we represent our sur-rounding space and the spatial properties of objects are determined by how we are able to act (see also the body specificity hypothesis for abstract concepts, Casasanto, 2009). For example, in Morgado, Gentaz, Guinet, Osiurak, and Palluel-Germain (2013)’s experiment, par-ticipants were asked to report the distance they saw between their hand and a target located behind a transpar-ent barrier of varying width. Before each evaluation, par-ticipants were asked to mentally simulate the action of grasping the target with their hand by reaching it around the barrier. Results analysis showed that participants’ estimations of the target distances were higher after they had to imagine grasping the target around wider barriers than narrower ones (for similar results with genuinely performed actions, see Kirsch & Kunde, 2012, 2014). This result highlights that variations on the potential action range entail correlative changes on how we repre-sent the target position in space.

ACT, as TEC, predict that when we perceive a target, the motor codes activated by the task execution are inte-grated (i.e., linked) to the perceptive representation of the target. However, ACT highlight something more about this integration: it leads to a distortion of the target location. Interestingly, this distortion can be predicted by the variations of the action spatial properties. The characteristics of the imagined action entail correlative changes on the spatial representation of the target. We can reasonably expect that the movement-effect integra-tion at stake during the ideomotor two-stage procedure leads to the same kind of correlative distortion of the effect representation. The properties of action influence how we localise action effects in space (i.e., depth, dis-tance, height, horizontal position) and how we represent its spatial properties (i.e., size, width, depth). We thus suggest that integration in the ideomotor bidirectional association is something more than a linkage process between action and perception. We hypothesise that movement properties have a correlative influence on the spatial representation of their effects.

The present study

ACT invite ideomotor theories to consider that movement-effect association probably impacts the effect representa-tion. Moreover ACT bring us to expect that changes in the spatial movement properties (e.g., direction) entail correl-ative changes in the spatial representation of the effect. To examine this claim, in our experiments, we associated non-spatialised auditory effects to specific movements during an ideomotor acquisition phase. We expected movement directions (rightward, leftward, forward, and backward movements) to entail correlative changes on how the audi-tory effects were localised in space.

Before starting any investigation, we conducted a pre-experiment as a replication of Kunde (2001)’s experiment to control that the R-E compatibility procedure we used in our following-up studies really involved an ideomotor mechanism. As the results of this pre-experiment were in line with our expectations and with Kunde (2001)’s results, we applied the corresponding mapping condition of the pre-experience to both our experiments—as a condition that involved an ideomotor mechanism. For each experi-ment, we used a pre-test-post-test experimental design divided in four phases: set-up, pre-test, acquisition, and post-test phases.

The goal of the first experiment was to test our main hypothesis concerning the correlative influence of move-ment properties on effect representation. The goal of the second experiment was to consolidate the interpretation of the obtained results by investigating possible alternative explanations of the first experiment findings.

The pre-experiment: a replication of Kunde (2001)’s experiment

The aim of this pre-experiment was first to replicate Kunde (2001)’s results: a spatial R-E corresponding mapping leads to faster response times than a R-E non-correspond-ing mapping. Second, our goal was to have a R-E compat-ibility procedure (i.e., the corresponding mapping condition) involving an ideomotor mechanism, to test our main hypothesis in the following experiments.

Based on Kunde (2001)’s procedure, our pre-experi-ment design differs in that the procedure was made to increase the anticipation of the response effects: we chose a paradigm in which events were perfectly predictable. In that respect, all our conditions were always fixed, such as in Hoffmann, Sebald, & Stöcker (2001). This means that in contrast to classic ideomotor procedures, our design does not allow testing response selection per se, because the decision concerning which movement to perform could be made before the stimulus is presented. Indeed, our main investigation aims at studying effect representation and not response selection. We chose this predictable paradigm to make the task as simple as possible for the participants.

4 Quarterly Journal of Experimental Psychology 00(0)

Moreover, by maintaining the sequences of response-effect associations constant, we ensured that only the direction of the movement changed between experimental conditions in the following-up experiments.

Method

Participants. In all, 40 students of the University of Paris-Nanterre (22 males, participants mean age 21.73 years old, SD: 2.84) took part in the pre-experiment. All had normal or corrected-to-normal vision and audition. Two experi-menters managed 20 participants each in a randomised fashion. Informed consent was obtained from all partici-pants included in the study.

Apparatus and material. The whole experimental setup was placed on a 65 x 120 cm2 desk. The experiment took place on an ASUS X53 S laptop (15.6-inch screen, running Win-dows XP, Core i3 processor), and was programmed with Python v2.7.11 software (https://www.python.org/). The laptop screen was placed at 1 cm from the upper edge of the desk and centred on its length. Participants were given a Sennheiser 380 PRO audio headset to wear.

Two kinds of stereo tones were used: a high-pitched pure tone (1,777 Hz) and a low-pitched pure tone (110 Hz), each with 500 ms duration, a digital resolution of 32bits and digitised at 48,000 samples/s. These tones were pre-sented through the audio headset either to the left ear or to the right ear of the participant. We therefore used four auditory tones: the low-pitched tone heard in the left ear, the high-pitched tone heard in the left ear, the high-pitched tone heard in the right ear, and the low-pitched tone heard in the right ear. These tones have been designed with Audacity v.2.1.2 software (http://www.audacityteam.org/).

Visual scenes were displayed on the laptop screen, all representing a ghost symbol of 2.8 x 1.8 cm2, located in the centre of the screen on a white background. Half of the ghost symbols were coloured in blue, the other half were coloured in yellow. These stimuli were preceded by a fixa-tion cross of font size 48 and of 500 ms duration.

The answer apparatus has been built with the device Makey-Makey®. It comprised three squared 4 x 4 cm2 but-tons. One, coloured in red, was located in the centre of the desk length with its centre 3 cm away from its lower edge. The other two, covered with aluminum, were aligned with the red button, on the X-axis with respect to the transverse plane of the subject, with their centres 5 cm away from the left and right edges of the desk. Both had a blue or yellow sticker placed in their centre. They were connected to the laptop and converted into response buttons by the Makey-Makey device.

Procedure. The subjects’ task was to match each ghost symbol with the response button presenting the same col-our. Ghost symbols were displayed in the centre of the

screen and were preceded with the fixation cross. Subjects used both their hands to answer, the left hand for the left side and the right hand for the right side. The yellow colour was associated with the left side and the blue colour with the right side; this organisation was inverted for half of the participants. They were told to start from the red button located at the centre, go and press the button matching the colour of the ghost, and come back to the centre. They had to answer as quickly as possible from the onset of visual stimuli. The screen was cleared after each subject’s response. Response buttons each triggered an auditory effect displayed in the headset. As in Kunde (2001)’s experiment, we designed two experimental conditions: an R-E spatial corresponding mapping and a non-correspond-ing mapping. For the corresponding mapping, tones heard in the left ear were the effect of the left response button, and tones heard in the right ear were the effects of the right response button. For the non-corresponding mapping, tones heard in the right ear were the effects of the left response button, and tones heard in the left ear were the effects of the right response button. After an intertrial interval of 1,000 ms, the fixation cross signalling the fol-lowing trial was displayed.

To measure the level of influence of a R-E compatibil-ity on movement selection according to the temporal evo-lution of the experiment, we designed 12 blocks as a within-participants factor for each condition (correspond-ing vs. non-corresponding mapping). The first block was labelled as training. Each block in the corresponding map-ping condition followed this systematic pattern: (1) A fixa-tion cross was displayed, followed by a yellow ghost. The subject made a movement towards the left and received as a consequence a low-pitched tone in his left ear. (2) After coming back to the centre button, a fixation cross was dis-played, followed by a yellow ghost. The subject made a movement towards the left and received as a consequence a high-pitched tone in his left ear. (3) After coming back to the centre button, fixation cross was displayed again, fol-lowed by a blue ghost. The subject made a movement towards the right and received as a consequence a high-pitched tone in his right ear. (4) After coming back to the centre button, a fixation cross was displayed, followed by a blue ghost. The subject made a movement towards the right and received as a consequence a low-pitched tone in his right ear. The blocks in the non-corresponding condi-tion followed the same order.

We recorded response times in milliseconds for each response (time between the onset of stimuli and button press). A power analysis (conducted via G*Power Software, Faul, Erdfelder, Lang, & Buchner, 2007, with Cohen’s [1988] recommendations), which assumed a medium effect size of 0.25 for the analysis of variance (ANOVA) with one between-subjects factor and one within-subjects factor (11 levels of signs as repeated meas-ures), which would not interact together, indicated that a

Lestage et al. 5

total of 16, 14 and 14 participants were required, respec-tively, to have a 90%, 85%, or 80% power (a minimum required by Cohen [1988]) of detecting a significant effect at p value of 0.05. In all, 20 subjects performed the corre-sponding condition and 20 subjects performed the non-corresponding condition (for a total of 40 participants).

Results

Responses with RTs below 200 ms and above 2,700 ms were discarded as outliers (>3.3% of all responses). We performed an ANOVA on the RTs, with R-E compatibility (corresponding vs. non-corresponding mapping) as a between-participants factor and Blocks (2-12, excluding the first training block) as a within-participants factor. The mean RTs for the corresponding versus non-corresponding mapping were 619 versus 840 ms. The ANOVA of RTs revealed that this difference was significant F(l, 38) = 25.3, p < .01, ηG² = 0.34, ηp² = 0.4. Subjects were faster with the corresponding mapping than with the non-corresponding mapping. The effect of the Blocks factor was also signifi-cant, F(10, 380) = 6.2, p < .01, ηG² = 0.04, ηp² = 0.14. This means that RTs became faster and faster for each block. The interaction between R-E compatibility and Blocks factors was not significant (p > .1). See Figure 1 for an illustration of these results.

Discussion

One of the aims of this pre-experiment was, first, to repli-cate Kunde (2001)’s results: a spatial R-E corresponding mapping leads to faster response times than R-E non-cor-responding mapping. The obtained results are in line with this assumption: when the location of a subject’s response corresponded to the location of its auditory effect, responses were faster than when these locations did not correspond.

The principal goal of this pre-experiment was to build an experimental procedure involving an ideomotor mecha-nism, to test our main hypothesis in the following experi-ments. Results obtained with the corresponding mapping condition confirm that this goal was achieved. We were thus able to use this condition in the following experiments as an acquisition phase. We tested in which extent the movement spatial properties (i.e., direction) can entail cor-relative changes in the effect spatial representation.

Experiment 1

In this experiment, we meant to investigate how move-ment properties within an ideomotor bidirectional associa-tion impact the effect spatial representation. This investigation answers Shin, Proctor, and Capaldi (2010)’s remark that ideomotor theories should turn their interest on how the properties of action impact perception. Moreover, our predictions are in line with ACT claim according to which changes in the spatial properties of a movement (e.g., direction) entail correlative changes in the spatial representation of perceptive events surrounding the move-ment. Considering that our interest was to investigate the bidirectional ideomotor association between movements and effects, the perceptive event we focused on was the movement effect.

We used a pre-test-post-test experimental design divided in four phases: set-up, pre-test, acquisition, and post-test phases. Only the acquisition phase varied between two different conditions. For the first condition, partici-pants performed movements along the transverse axis (i.e., leftward and rightward movements). In the second condi-tion, participants performed movements along the ante-rior–posterior axis (i.e., forward and backward movements). The acquisition phase was drawn on the R-E corresponding mapping condition of the pre-experiment. See Table 1 for an illustration of the different variations of the acquisition phase in both conditions.

In the next section and for a better understanding, we present the description of the material and the procedure separately for each phase of the experiment. See Figure 2 for an illustration of the procedure followed in the set-up and pre- and post-test phases.

Method

Participants. A total of 96 students of the University of Paris-Nanterre (43 males, participants mean age 19.8 years old, SD 1.97) took part in this experiment. All had normal or corrected-to-normal vision and audition. Four experi-menters managed 24 participants each in a randomised fashion. Informed consent was obtained from all individ-ual participants included in the study.

Apparatus. The experimental setup was the same as in the pre-experiment.

Figure 1. Results of the pre-experiment: RT in ms as a function of corresponding vs. non-corresponding R-E mappings and blocks. The bars represent standard errors.

6 Quarterly Journal of Experimental Psychology 00(0)

Set-up: the explicit tone-location first phase

Material. Four visual scenes were presented on the laptop screen. Each represented a circle of 1.5 cm diameter filled in green, on a white background. In the first visual scene, the circle was located in the lower-left space of the screen; in the second scene, the circle was located in the upper-left space; in the third scene, the circle was located in the upper-right space; and in the fourth scene, the cir-cle was located in the lower-right space. If brought together, the four circles’ centres were located on the four corners of a 6.5 x 10.5 cm2 rectangle—the centre of the screen being the centre of the rectangle. The location of each circle was coupled with a tone in a compatible manner (see Rusconi, Kwan, Giordano, Umiltà, & But-terworth, 2006, and Kunde, 2001, on spatial correspond-ence between pitches and location), as specified in the

following section. The four auditory stimuli were the same as in the pre-experiment.

Procedure. Participants were placed in front of the laptop screen, on which instructions were displayed (see Figure 2). They were asked to memorise couplings between loca-tions of visual stimuli (represented by the green circles) and locations of tones (heard through the audio headset). The sequence of visual-auditory couplings presented to the subject was (1) a low-pitched tone heard in the left ear was coupled with the circle located in the lower-left space of the screen, (2) a high-pitched tone heard in the left ear was coupled with the circle located in the upper-left space of the screen, (3) a high-pitched tone heard in the right ear was coupled with the circle located in the upper-right space of the screen, and (4) a low-pitched tone heard in the right ear was coupled with the circle located in the

Figure 2. Illustration of the procedure for the set-up phase, and the pre- and post-tests phases, for both experiments.

Table 1. Variations in the acquisition phase for both experiments.

Experiment 1 Experiment 2

Horizontal outward

Forward-backward

Movement with effect

Movement with no-effect

No-movement with effect

Movement

Horizontal outward Horizontal outward Key press

Movement effect Yes Yes Yes No Yes

Lestage et al. 7

lower-right space of the screen. Each circle was displayed for 1 s; the beginning of this display was synchronised with the beginning of the tone. After a first presentation, this sequence was repeated five times. We ensured that the four visual-auditory couplings had been properly memo-rised in the pre- and post-test phases. The pre-test followed the first set-up phase, while the post-test followed the acquisition phase. We will present them both last.

Acquisition: the R-E procedure third phase

Material. Ghost stimuli and response apparatus were simi-lar to the ones used in the pre-experiment. The few changes are justified by the fact that the forward–backward move-ments condition needed some adaptation to be compared to the horizontal outward movements condition.

Ghost symbols were the same size and shape as in the pre-experiment. They were white with a black contour line and displayed in their centre digits of size 16. The first ghost symbol displayed digit 1, the second displayed digit 2, the third displayed digit 3, and the fourth displayed digit 4. The answer apparatus for both conditions comprised five squared 4 x 4 cm2 buttons.

For the horizontal outward movements group, a red but-ton was located in the centre of the desk length and at 8 cm from its lower edge. The other four were covered with alu-minum. If considering the centre of the red button as a ref-erence point with 0 cm as X and Y coordinates, then the coordinates of the first button’s centre were -53 cm, -5 cm; the coordinates of the second button centre were -53 cm, 5 cm; the coordinates of the third button centre were 53 cm, 5 cm; and the coordinates of the fourth button centre were 53 cm, -5 cm. If brought together, the four button centres were thus located on the four corners of a 106 x 10 cm2 rectangle. Digits of font size 72 and blue colour were dis-played in the centre of the buttons, corresponding to the digits displayed on the screen: 1 for the button at the lower-left space, 2 for the button at the higher-left space, 3 for the button at the higher-right space, and 4 for the button at the lower-right space. Their function was to indicate the sub-ject where to press according to the instructions.

For the forward–backward movements group, one red-coloured button was located at the centre of the apparatus, in the centre of the desk length and at 19 cm from its lower edge. The other four were covered with aluminum. If con-sidering the centre of the red button as a reference point with 0 cm as X and Y coordinates, then the coordinates of the first button centre were -3 cm, -16 cm; the coordinates of the second button centre were -3 cm, 16 cm; the coordi-nates of the third button centre were 3 cm, 16 cm; and the coordinates of the fourth button centre were 3 cm, -16 cm. If brought together, the four button centres were thus located on the four corners of a 32 x 6 cm2 rectangle. Digits of font size 72 and blue colour were displayed in the centre of the buttons, corresponding to the digits displayed on the screen

(as in the horizontal outward movements group) to indicate the subject where to press according to the instructions.

Procedure. All participants were told that they were going to perform a distractive task consisting in a game where they had to “kill ghosts” as fast as possible (see Table 1). We used the same procedure as in the corresponding mapping condi-tion of the pre-experiment procedure. The only difference was that participants had to move their hands to one of the four buttons positioned in front of them (versus two buttons in the pre-experiment). For the first group, pressing a button corresponded to initiating a forward or backward move-ment. For the second group, pressing a button corresponded to initiating a horizontal outward movement to the left or to the right. Subjects used the left hand for the left buttons in response to ghosts with Symbols 1 and 2, and the right hand for the right buttons in response to ghosts with Symbols 3 and 4. They were told to start from the red button located in the centre, go and press the button matching the digit of the ghost, and return back to the centre. In line with Koch and Hoffmann (2000)’s procedure, the acquisition phase thus followed this pattern: (1) A fixation cross was displayed during 500 ms, followed by the “1” ghost. The subject made a movement with his left hand and received as a conse-quence a low-pitched tone in his left ear. He came back to the centre button. (2) A fixation cross was displayed, fol-lowed by the “2” ghost. The subject made a movement with his left hand and received as a consequence a high-pitched tone in his left ear. He came back to the centre button. (3) A fixation cross was displayed, followed by the “3” ghost. The subject made a movement with his right hand and received as a consequence a high-pitched tone in his right ear. He came back to the centre button. (4) A fixation cross was dis-played, followed by the “4” ghost. The subject made a movement with his right hand and received as a conse-quence a low-pitched tone in his right ear. He came back to the centre button. This sequence was repeated 12 times.

Implicit measure of the effect representation: pre- and post-test second and fourth phases

In these phases, we primed participants with the same sequence of four tones previously heard during the set-up phase and asked them to affix crosses on answer sheets according to the visual-auditory couplings they had memorised.

Material. To transcribe the locations of the tones, partici-pants used a sharp pencil on blank sheets of A4 paper; one sheet by transcribed location and by participant, namely eight sheets per subject. These sheets were presented in landscape orientation inside an opened cut box (21.5 x 29.5 x 2 cm3) fixed near the participant, at 5 cm from the lower edge of the desk and centred on its length. The box was taken out of the desk during the acquisition phase.

8 Quarterly Journal of Experimental Psychology 00(0)

Procedure. Pre-test and post-test phases were the same in all aspects (see Figure 2). The pre-test (second) phase followed the set-up phase, and the post-test (fourth) phase followed the acquisition phase. Subjects were given a pencil and were informed that they were going to hear the same sequence of four tones presented during the set-up phase. After each heard tone, their task was to affix a cross on the sheet in front of them according to the four visual-auditory associations they had memorised. Participants were not given any time limit to realise the task. For each cross affixed by the participant, the sheet was collected by the experimenter to avoid any possible influence of the preceding answers. At the end of the pre-test phase, they were told that this task would be followed by a distractive one (the acquisition phase), after which they would have to redo the exact same task and therefore recall the couplings again.

Data registered were first the X and Y coordinates in centimetres of the centre of each cross obtained during the pre- and post-test phases, according to an orthonormal coordinate system whose origin (the point where X = 0 and Y = 0 cm) corresponded to the centre of the A4 sheet in landscape orientation. To allow the comparison of the dif-ferent locations of crosses, only the absolute values of the obtained numbers were considered. A power analysis (conducted via G*Power Software, Faul et al., 2007 with Cohen’s recommendations, 1988), which assumed a medium effect size of 0.25 for the ANOVA with one between-subjects factor (2 levels of signs) and two within-subjects factors (2 levels of signs as repeated measures), which interact together, indicated that a total of 30, 26, and 24 participants were required, respectively, to have a 90%, 85%, or 80% power (a minimum required by Cohen [1988]) of detecting a significant effect at p value of 0.05. In all, 48 subjects performed the horizontal outward condi-tion and 48 performed the forward–backward condition (a total of 96 participants).

We also registered the discrepancy in centimetres between the locations of crosses on the sheet at the pre-test and at the post-test on both axes. The discrepancy was thus obtained by subtracting the absolute values found during the pre-test to the absolute values found during the post-test, for X and Y axes. A power analysis (conducted via G*Power Software, Faul et al., 2007 with Cohen’s recommendations, 1988), which assumed a medium effect size of 0.25 for the

ANOVA with one between-subjects factor (2 level of signs) and one within-subjects factor (2 levels of signs as repeated measures), which interact together, indicated that a total of 44, 38, and 34 participants were required, respectively, to have a 90%, 85%, or 80% power (a minimum required by Cohen [1988]) of detecting a significant effect at p value of 0.05. A total of 48 subjects performed the horizontal out-ward condition and 48 performed the forward–backward condition (a total of 96 participants).

Results

No error was made during the pre-test and post-test parts of the experiment. All participants drew the crosses where it was expected according to the heard tone: low-left space of the sheet for low-pitched left tone, high-left space for high-pitched left tone, high-right space for high-pitched right tone, and low-right space for low-pitched right tone.

To test our hypothesis, we used the absolute measures of the locations of crosses drawn by the subjects during the pre- and post-test phases to conduct a first analysis with a mixed ANOVA. Measure (pre-test, post-test) and Axis (X, Y) factors were within-participants measures and move-ment direction (horizontal outward vs. forward-backward) was a between-participants factor. Results are summarised in Table 2.

The analysis did not reveal a main effect of the Movement direction factor (p = .66): the two conditions globally led to similar tone localisation. We obtained a main effect of the Measure and Axis factors, respectively, F(1, 94) = 140, p < .01, ηG² = 0.3, ηp² = 0.6; F(1, 94) = 301, p < .01, ηG² = 0.7, ηp² = 0.8. Participants affixed their crosses further from the sheet centre during the post-test (mean 7.1 cm) than during the pre-test (mean 6.4 cm). Crosses were globally positioned further from the centre of the sheet on the X-axis (7.9 cm) than on the Y-axis (5.6 cm). The most important result was the significant interaction between Movement direction and Axis factors, F(1, 94) = 20.3, p < .01, ηG² = 0.1, ηp² = 0.2. On the Y-axis, crosses were drawn further from the sheet centre (p < .01) for the forward–backward condition (5.8 cm) than for the horizontal outward condition (5.3 cm). Reversely, on the X-axis crosses were drawn further from the sheet centre (p < .01) for the horizontal outward condition (8.3 cm) than for the forward–backward condition (7.5 cm).

Table 2. Summary of results for Experiment 1: absolute values of cross positions in centimetres.

Condition X-axis Y-axis p

Pre-test Post-test Pre-test Post-test

M SD M SD p M SD M SD

Horizontal outward 7.8 2.1 8.8 1.3 <.001 5.1 2.3 5.5 1.5 <.001Forward–backward 7.3 1.8 7.8 1.2 <.001 5.4 2 6.2 1.2 <.001

Lestage et al. 9

For a better support of this last result and to compare the two conditions to each other on a statistical level, we conducted a second analysis based on the discrepancies between cross locations from the pre-test and cross loca-tions from the post-test on both axes. The measured values were analysed with a mixed ANOVA: Axis (X, Y) factor was a within-participants measure and Movement direc-tion (horizontal outward movements and forward-back-ward movements) was a between-participants factor. Results are summarised in Table 3.

The analysis revealed no main effect of the Movement direction factor (p = .75): discrepancy between the loca-tions of crosses during pre- and post-test was globally the same for the horizontal outward movements condition (0.70 cm) and for the forward–backward movements con-dition (0.66 cm). Analysis also revealed no significant effect of the Axis factor (p = .2): discrepancy between the locations of crosses during pre- and post-test was globally alike for the X-axis (0.75 cm) and for Y-axis (0.62 cm). The important result obtained here was the significant interac-tion between Movement direction and Axis factors, F(1, 94) = 16.2, p < .01, ηG² = 0.07, ηp² = 0.15. Post hoc analyses revealed that the difference of discrepancy for the horizon-tal outward movements condition between X-axis (0.97 cm) and Y-axis (0.42 cm) was significant, F(1, 94) = 13.9, p < .01. Similarly, for the forward-backward movements condition the difference of discrepancy between X-axis (0.52 cm) and Y-axis (0.81 cm) was quasi-significant F(1, 94) = 3.9, p = .051. The most interesting result was that the significant difference of discrepancy on the X-axis between the horizontal outward movements (0.97 cm) and the for-ward-backward movements (0.52 cm), F(1, 94) = 6.9, p < .02. Similarly, the difference of discrepancy on the Y-axis between the horizontal outward movements (0.42 cm) and the forward-backward movements (0.81 cm) conditions was significant F(1, 94) = 8.4, p < .01. When participants performed horizontal outward movements during the acquisition phase, their tone localizations became enlarged on the X-axis more than on the Y-axis.

Reversely, when participants performed forward and back-ward movements during the acquisition phase, their tone localizations became enlarged on the Y-axis more than on the X-axis. See Figure 3 for an illustration of these results.

Discussion

The goal of this first experiment was to investigate how movement properties within an ideomotor bidirectional association impact the effect spatial representation. We suggest that changes in the spatial properties of a move-ment (e.g., direction) entail correlative changes in the spa-tial representation of the movement effects. We investigated the influence of lateralised vs. backward-forward arms movements on the spatial representation of sensory effects.

We conducted two analyses on the tone locations reported by the participants, one on absolute values and another one on discrepancies between the locations of crosses during pre-test and during post-test. This allowed us to assess changes in tone locations reported by the par-ticipants. Both analyses revealed an interaction between the Movement direction and the Axis factor. This interac-tion indicates that when participants performed forward–backward movements they were subsequently more likely to locate the crosses more up and down on the sheets dur-ing the post-test phase than during the pre-test phase. Conversely participants who performed horizontal out-ward movements were more likely to locate their crosses more to the right and left edges of the sheets during the post-test phase than during the pre-test phase. In sum, hori-zontal outward movements enlarged the spatial representa-tion of action effect more on its horizontal spatial dimension (X axis) than on its vertical one (Y axis). Reversely forward–backward movements enlarged its

Table 3. Summary of results for the two experiments: discrepancies in centimetres.

Condition On X-axis On Y-axis p

M SD M SD

Experiment 1Horizontal outward 0.98 1 0.42 0.6 <.001Forward-backward 0.52 0.7 0.81 0.7 .05

Experiment 2Movement with effect 0.97 0.8 0.34 0.5 <.001Movement with no-effect 0.42 0.8 0.38 0.6 .8No-movement with effect -0.03 0.8 0.12 0.5 .26

Figure 3. Mean discrepancies (cm) for X-axis and Y-axis for Experiment 1 (horizontal outward and forward-backward movements). The bars represent standard errors.

10 Quarterly Journal of Experimental Psychology 00(0)

spatial representation more on its vertical dimension (Y axis) than its horizontal one (X axis). This result supports our hypothesis according to which the movement-effect integration during the ideomotor two-stage procedure leads to a distortion of the localisation of the effects. Moreover, these results fall in line with our main hypoth-esis about the correlative influence of movement proper-ties on the effect representation: horizontal outward movements enlarge cross locations along the horizontal axis, whereas backward–forward movements enlarge cross locations in a backward and forward direction.

Nevertheless, an alternative explanation has to be men-tioned. It is possible that the observed shift in effect locali-sation is not due to the specific spatial properties of movements performed during the acquisition phase. Changes in effect representation could be due to a simple pre-test-post-test impact: subjects shifted the location dur-ing the post-test because it was the second time they were performing the task. To ensure that spatialised movements were really the impacting factor in our results, in a follow-ing experiment we tested a condition where tone effects were associated with simple finger key presses, as opposed to large movements with direction in space (no-movement with effect condition of experiment 2).

The other results obtained with the absolute values analysis also have to be commented. This analysis revealed a significant main impact of the Axis factor: the locations of crosses were globally more distant from the sheet centre for X-axis than for Y-axis. This impact is likely due to the specific feature of our answer sheets format—an A4 paper presented in landscape orientation.

More interestingly, the absolute values analysis revealed a main effect of the Measure factor. Participants affixed their crosses further from the sheet centre during the post-tests than during the pre-tests. This means that participants globally enlarged cross locations on both X and Y axes after perform-ing the acquisition phase. One way to interpret this result is to make reference to what ACT suggests: action properties are integrated to the perceptive process involved during a locali-sation task. Based on this assumption, it is possible that move-ment direction became integrated to more general perceptive processes than those strictly running between the movement and the effect during the acquisition phase. In other words, it might be possible that our results do not rely on an ideomotor bidirectional association, but simply on the movements per-formed during the acquisition phase. Indeed, in both condi-tions of the experiment, each movement implied an increase of the distance between the subjects’ hands. It is thus possible that this increase became integrated to all the perceptive events surrounding the movement—not just the effect. This idea is based on experimental works showing that the body influences size perception and can thus impact perceptive judgements (see Linkenauger, Proffitt, & Witt, 2011; Stefanucci & Geuss, 2009). To test this alternative explana-tion, and to ensure that the impact of movement is really depending on the bidirectional association, we tested an

experimental condition in which movements during the acquisition phase were not associated to any effect (move-ment with no-effect condition of Experiment 2). Tones were not heard during the acquisition phase anymore. If partici-pants still enlarge tone locations during the post-test, we will be able to confirm the general impact of movement on per-ceptive processes. This would mean that movement is inte-grated to more general perceptive process than those strictly running between the movement and the effect. Nevertheless, this result would not rule out a possible influence of the bidi-rectional ideomotor association. To test this influence, we compared the movement with no-effect condition to a move-ment with effect condition equivalent to the corresponding mapping condition of the pre-experiment.

Experiment 2

The goal of this experiment was to test in which extent the results observed in Experiment 1 were the consequence of the integration of movement direction with the effect rep-resentation. In that end, Experiment 2 was built first to ensure that the results observed in Experiment 1 were due to the impact of movement and not to a simple pre-test-post-test impact. Second, Experiment 2 tested in which extent movement was integrated to the effect representa-tion, as opposed to being integrated to more general per-ceptual processes.

Experiment 2 was designed on Experiment 1. This time we compared three experimental conditions. The first movement with effect condition was similar to the horizontal outward movement condition of Experiment 1 and equivalent to the corresponding mapping condi-tion of the pre-experiment. For the second movement with no-effect condition, the difference was that move-ments did not produce any auditory effect. In the third no-movement with effect condition, simple finger key presses replaced the arm movements performed to press the response buttons. See Table 1 for an illustration of the different variations of the acquisition phase for the three conditions.

Method

Participants. A total of 108 students of the University of Paris-Nanterre (35 males, participants mean age 20.65 years old, SD 2.28) took part in the experiment. All had normal or corrected-to-normal vision and audition. Three experi-menters managed 36 participants each in a randomised fashion. Informed consent was obtained from all individ-ual participants included in the study.

Apparatus and stimuli. They were the same as in the Exper-iment 1 for the set up first phase and for the pre- and post-test second and fourth phases. They were the same as in the corresponding mapping condition of the pre-experiment for the acquisition third phase.

Lestage et al. 11

Procedure. We used the same procedure divided in four phases as in Experiment 1. Set-up, pre-test and post-test phases were the same as in Experiment 1 (see Figure 2).

During the acquisition phase, a first group followed the corresponding mapping condition of the pre-experiment procedure. Four auditory effects were associated to two horizontal outward movements (rightward or leftward). The two tones (high and low pitches) heard in the left ear were the effect of leftward movements in response to blue ghost stimuli, and the two tones (high and low pitches) heard in the right ear were the effect of rightward move-ments in response to yellow ghost stimuli (movement with effect condition). Colours on response buttons were coun-terbalanced. For a second group, the procedure only dif-fered in that spatialized movements performed as responses to ghost stimuli were replaced by simple key presses (no-movement with effect condition). In this condition, blue and yellow stickers were displayed on the centre of the “B” and “N” keys of the laptop keyboard—the space between these keys being at the centre of the keyboard length—to indicate the subject where to press according to the displayed ghost colour. For a third group, the proce-dure only differed from the movement with effect condition in that participants received no auditory effect after their responses (movement with no-effect condition). Instead, the rightward and leftward movements performed by the subjects were followed by a 500 ms silence (see Table 1).

We registered, as in Experiment 1, the discrepancies in centimetres between the pre-test and the post-test cross locations on both axes. A power analysis (conducted via G*Power Software, Faul et al., 2007 with Cohen’s recom-mendations, 1988), which assumed a medium effect size of 0.25 for the ANOVA with one between-subjects factor and one within-subjects factor (2 levels of signs as repeated measures), which would interact together, indicated that a total of 54, 48, and 42 participants were required, respec-tively, to have a 90%, 85%, or 80% power (a minimum required by Cohen [1988]) of detecting a significant effect at p value of 0.05. In all, 36 subjects performed the move-ment with effect condition, 36 performed the movement with no-effect condition, and 36 performed the no-move-ment with effect condition (a total of 108 participants).

Results

As for experiment 1, no error was made during the pre- and post-tests parts of the experiment. To test our hypoth-esis, we conducted an analysis based on the discrepancy between the locations of crosses from the pre-test to the locations of crosses from the post-test on both axes. The measured values were analysed with a mixed ANOVA: Axis (X, Y) was a within-participants measure and type of acquisition (movement with effect, movement with no-effect, no-movement with effect) was a between-partici-pants factor. Results are summarised in Table 3.

The ANOVA revealed a main effect of the type of acquisition factor, F(2, 105) = 10.6, p < .01, ηG² = 0.12, ηp² = 0.17. The mean discrepancy (both axes combined) between pre- and post-tests cross locations was statisti-cally different for the three conditions. Least significant difference (LSD) Post hoc analysis revealed that the differ-ence of discrepancy between the movement with effect condition (0.65 cm) and the movement with no-effect con-dition (0.4 cm) tended to statistical significance (p < .06). The discrepancy between the movement with effect condi-tion (0.65 cm) and the no-movement with effect condition (0.05 cm) was statistically different (p < .01). The discrep-ancy between the movement with no-effect condition (0.4 cm) and the no-movement with effect condition (0.05 cm) was statistically different (p < .01). We also observed a main effect of the Axis factor, F(1, 105) = 5.2, p < .03, ηG² = 0.02, ηp² = 0.05: the discrepancy was larger on the X-axis (M = 0.45 cm) than on the Y-axis (M = 0.28 cm). At last, the most interesting result was the interaction observed between Axis and type of acquisition factors, F(2, 105) = 9.9, p < .01, ηG² = 0.06, ηp² = 0.16; see Figure 4. LSD post hoc analysis revealed a significant dif-ference between X (0.97 cm) and Y axes (0.34 cm) for the movement with effect condition (p < .01), but no signifi-cant difference between both axes for the movement with no-effect condition (X = 0.42 cm, Y = 0.38 cm; p = .8) nor for the no-movement with effect condition (X = -0.03 cm, Y = 0.12 cm; p = .24).

Discussion

The goal of this second experiment was to investigate in which extent the influence of movement on effect representation is related to an ideomotor bidirectional association. Indeed, as said above, it is possible that the

Figure 4. Mean discrepancies (cm) for X-axis and Y-axis for the three conditions of experiment 2. The bars represent standard errors.

12 Quarterly Journal of Experimental Psychology 00(0)

discrepancy in cross locations obtained with Experiment 1 was the consequence of a pre-test-post-test impact, thus discarding any influence of the bidirectional association. Moreover, it is possible that the movements performed during the acquisition phase were integrated to more gen-eral perceptive processes than those operating between movement and effect.

We built three conditions: movement with effect, move-ment with no-effect, and no-movement with effect condi-tions. We conducted an analysis on the discrepancy between pre- and post-test tone locations reported by the participants. Regarding the goals of this experiment, the first important result concerns the movement with effect condition. The significance of the type of acquisition fac-tor revealed that this condition was the one entailing the largest shifts in effect localisation. The significant interac-tion between Axis and Type of acquisition factors revealed that this shift was correlative to the performed move-ments: participants enlarged their tone spatial representa-tion on the X-axis more than on the Y-axis. These results confirm and reinforce the previous results obtained in Experiment 1: being subject to a bidirectional ideomotor association leads to representational changes according to the movement axis.

On the other hand, the same factors also revealed that the movement with no-effect condition entailed significant changes on the effect representation as well. Participants enlarged cross locations after the acquisition phase on both axes. Movements thus had an influence on tone localiza-tions even when these tones were not the effect of move-ments. As mentioned above, this could be the sign that the integration of movement does not only happen with effects, but also with other perceptive processes involved. Nevertheless, the significance of the Axis factor and the significant interaction between Axis and Type of acquisi-tion factors also revealed that only the movement with effect condition led to specific changes in the spatial repre-sentation of the effect. Only in this condition did move-ment specific properties (i.e., movements constrained on the X-axis) reflect on the effect representation (tones were located as more enlarged on the X axis than on the Y-axis). In contrast, the movement with no-effect condition led to non-specific changes on the spatial representation of the auditory items. Indeed, we did not observe any difference in the spatial representation of the tone location between X and Y axes. Taken together, these results show that 1/ movement properties are integrated to general perceptive processes, as suggested by ACT and 2/ movement direc-tion has a specific influence on the spatial representation of the action effect.

Finally, result analysis did not reveal any shift on the tone location in the no-movement with effect condition. Crosses were localised at the same place after the acquisition phase than before—the discrepancy had a null value. This result allows ruling out the possibility of a pre-test-post-test

impact: subjects did not shift cross locations because the task was performed for the second time. This helps confirm-ing the dependence of the tone localisation shift on spatial-ized movements.

General discussion

In this article, we investigated how the spatial properties of movements affect the spatial representation of movement effects. In line with Gibson’s (1966) ecological theory, ideomotor theories allege that action and perception share a common substrate as well as a bidirectional influence between them. Perceived events (perceptions of stimuli and action effects) and to-be-produced events (actions) are equally represented so that the activation of a specific motor code can be predicted by the activation of a specific corresponding perceptive code, and reciprocally. Following Shin, Proctor, and Capaldi (2010)’s critic, we investigated how the properties of action impact perception in an ideo-motor bidirectional association. Until now, the few experi-ments having taken interest in this matter only referred to perception of stimuli preceding and triggering the action. We chose to take interest in action effect, thus investigat-ing the bidirectional association consequences on effect representation. In the experiments presented above, we highlighted that movement direction can affect how we represent the spatial components of its associated effect. Moreover, we showed that movement direction entails correlative changes in the effect representation. For exam-ple, a movement to the right associated to an auditory effect during an acquisition phase led participants to local-ise the effect more on the right after than before the acqui-sition phase. In sum, these results suggest that movement direction shapes the spatial properties of its sensory effects.

To the best of our knowledge, no experiment investi-gated the influence of movement properties on the spatial representation of the effect until now. We believe the reason of this shortage is that all ideomotor conceptions adhere to the fact that the first phase of the ideomotor mechanism is an integrative phase in which movement and effect are linked by a bidirectional association. This association is understood as a process by which two distinct events become connected or linked in a reciprocal manner. This is what Elsner and Hommel (2001) imply by saying that

given the temporal overlap of the activation of the motor code and the sensory pattern, the corresponding codes are integrated (i.e., linked with each other) so that activating one pattern on a latter occasion will lead to activating the other one, too. (p. 230)

Interestingly, the notion of integration refers here to this linkage process. Thereby, there is no mention of the possible consequences of this integration on how we represent each event after this process, as integration is no more no less than a connection between two events. This

Lestage et al. 13

conception suggests that movement and effect integration do not affect the representational content of each part.

Instead of a linkage process, we suggest that this inte-gration could reflect a unitisation process (Goldstone, 1998, 2000) by which a functional unit is built when a complex configuration arises. For Goldstone (2000) “a unit tends to be created if the parts that compose the unit frequently co-occur” (p. 98). Unitization leads to construct a unit which becomes more than the sum of its parts. According to Goldstone and Byrge (2015, p. 823), unitisa-tion in perception is akin to “chunking” in memory: differ-ent codes become integrated in one code that ends up having a different meaning than the sum of each part. In line with Goldstone’s view of integration, our study sug-gests that the ideomotor bidirectional association is more than a connection or a linkage process between movement and effects. Instead, it is better described as a unitisation process by which a unit is constructed (e.g., the action effect representation) which is more than the sum of its part. This conception of integration is more suitable to explain our experimental results than the linkage one. Indeed, if the process by which movement and effect are integrated together was a linkage process we should not have observed any change in the effect representation.

In further experiments, it would be interesting to test in which extent other movement properties such as range, strength, duration or fluency, could be integrated to the effect representation. Moreover, it would be interesting to explore more genuine “real life” movements such as in the domain of tool use. Recent theoretical models have pro-posed that tool-use actions could be based on an ideomotor principle (Osiurak & Badets, 2016, 2017). Based on these findings, we can predict that the tool properties would be integrated within the bidirectional association as part of movement properties. For example, the use of a long tool—such as a ruler—could expand the spatial represen-tation of effects even further in space than the arm move-ments in our experiments. In that respect, we could deepen our hypothesis that movement properties shape the spatial representation of the effect.

The shaping of the spatial representation of the effect by movement properties has another theoretical outcome concerning proximal versus distal information in TEC. Many studies investigating the ideomotor mechanism use already-spatialized visual or auditory action effects (Badets & Rensonnet, 2015; Pfister, Janczyk, Gressmann, Fournier, & Kunde, 2014; Thébault, Michalland, Dérozier, Chabrier, & Brouillet, 2018). Some experimental results justify this choice by showing that R-E compatibility effect arises within distal effects and not with proximal ones (Hommel, 1996; Osiurak & Badets, 2014; Pfister & Kunde, 2013). In this instance, distal effects refer to events objectively definable in the environment and perceived as the consequence of an action. Proximal effects refer to bodily, personal information given by sensory organs as

the consequence of an action (see Heider, 1926, 1930, and Brunswik, 1944). The studies previously mentioned seem to indicate that only distal effects are integrated in the bidi-rectional association. However, some recent studies high-light that proximal information actively participates to the selection of action (Pfister et al., 2014; Thébault et al., 2018). For example, in line with Kunde (2001, experiment 1), Thébault et al. (2018) showed that tactile vibrations with no spatial reference make it possible to observe ideo-motor compatibility effects. They conclude that proximal information (delivered by touch and proprioception) is part of the ideomotor mechanism. In fact, as it was initially proposed by James, ideomotor theory is always about the proximal information: before they get integrated by the association, action effects are nothing more and nothing less than raw sensory information. What our study sug-gests, in line with other studies about proximal informa-tion previously mentioned, is that the ideomotor bidirectional association is actually what gives distal refer-ence to a proximal information. Movements performed in space shape the space of their sensory effects.

This proposal is in line with the sensorimotor theory of perceptual experience (see O’Regan & Noë, 2001). This theory suggests that « perceiving is a way of acting. Perception is not something that happens to us, or in us. It is something we do » (Noë, 2004, p. 1). Action is what constitutes and defines the laws of perception: seeing is the ability to actively change our sensory impressions. In other words, the spatial properties of an object are shaped by sensorimotor contingency laws. In line with these assump-tions, we suggest that distal information is extracted from the sensorimotor contingency laws. It is not something coded as an internal representation from an external world, independently of our body. Regarding our experiments, we can presume that sensorimotor contingencies were formed by the repeated association between movements and audi-tory effects during the acquisition phase. These sensorimo-tor contingencies led to change the perceptual experience previously brought up during the set-up phase. This inter-pretation also echoes studies putting forward the crucial role of the body in perception (e.g., see the own-body-size effect of van der Hoort and Ehrsson (2014); or the distor-tion of perception by motion by Veto, Einhäuser, and Troje (2017)), as well as studies supporting the ACT; see Morgado & Palluel-Germain, 2015), and studies claiming that the body we have and how it moves determines how we think and how we perceive (the body-specificity hypothesis: see Casasanto, 2009 and Casasanto & Chrysikou, 2011).

To conclude, our study opens a new way of investiga-tion for ideomotor theory because, as said by Stock and Stock (2004), “all representatives of the ideomotor hypoth-esis expressed themselves rather unspecifically on the con-tents of the idea of a movement” (p. 186). All the considerations presented here suggest that all perceptual

14 Quarterly Journal of Experimental Psychology 00(0)

contents are built for and by movements, in the sense that a stimulus is nothing more and nothing less than a bidirec-tional action-effect construct.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Badets, A., & Rensonnet, C. (2015). Une approche idéomotrice de la cognition [An ideomotor approach of cognition]. L’Année Psychologique, 115, 591-635.

Brunswik, E. (1944). Distal focusing of perception. Size con-stancy in a representative sample of situations. Psychological Monographs, 56, 1-49.

Casasanto, D. (2009). Embodiment of abstract concepts: Good and bad in right and left handers. Journal of Experimental Psychology General, 138, 351-367.

Casasanto, D., & Chrysikou, E. G. (2011). When left is “right.” Motor fluency shapes abstract concepts. Psychological Science, 22, 419-422.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences. New York, NY: Routledge.

Craighero, L., Fadiga, L., Rizzolatti, G., & Umiltà, C. (1999). Action for perception. A motor-visual attentional effect. Journal of Experimental Psychology: Human Perception and Performance, 25, 1673-1692.

Deubel, H., & Schneider, W. X. (1996). Saccade target selection and object recognition: Evidence for a common attentional mechanism. Vision Research, 36, 1827-1837.

Elsner, B., & Hommel, B. (2001). Effect anticipation and action control. Journal of Experimental Psychology: Human Perception and Performance, 27, 229-240.

Fagioli, S., Ferlazzo, F., & Hommel, B. (2007). Controlling atten-tion through action: Observing actions primes action-related stimulus dimensions. Neuropsychologia, 45, 3351-3355.

Fagioli, S., Hommel, B., & Schubotz, R. I. (2007). Intentional control of attention: Action planning primes action-related stimulus dimensions. Psychological Research, 71, 22-29.

Faul, F., Erdfelder, E., Lang, A.-G., & Buchner, A. (2007). G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39, 175-191.

Gibson, J. J. (1966). The senses considered as perceptual sys-tems. Boston, MA: Houghton Mifflin Harcourt.

Goldstone, R. L. (1998). Perceptual learning. Annual Review of Psychology, 49, 585-612.

Goldstone, R. L. (2000). Unitization during category learning. Journal of Experimental Psychology: Human Perception and Performance, 26, 86-112.

Goldstone, R. L., & Byrge, L. A. (2015). Perceptual learning. In M. Matthen (Ed.), The Oxford handbook of philosophy of percep-tion (pp. 814–832). Oxford, UK: Oxford University Press.

Greenwald, A. G. (1970). A choice reaction time test of ideomo-tor theory. Journal of Experimental Psychology, 86, 20-25.

Harless, E. (1861). Der Apparat des Willens [The apparatus of the will]. Zeitschrift für Philosophie und Philosophische Kritik, 38, 50-73.

Heider, F. (1926). Ding und Medium. Symposion, 1, 109-157. Translated as Thing and medium in G.S.Klein (Ed.), Psychological Issues (pp. 1-34). New York: International Universities Press, Part 3, 1959.

Heider, F. (1930). The function of the perceptual system. Psychological Issues, 114, 371-384.

Herbart, J. F. (1825). Psychologie als Wissenschaft neu gegrün-det auf Erfahrung, Metaphysik und Mathematik [Psychology established as a science on experience, metaphysics and mathematics] (Zweiter, analytischerscher Teil). Königsberg, Russia: Unzer.

Hoffmann, J., Sebald, A., & Stöcker, C. (2001). Irrelevant response effects improve serial learning in serial reaction time tasks. Journal of Experimental Psychology: Learning, Memory, and Cognition, 27, 470-482.

Hommel, B. (1996). S-R Compatibility effects without response uncertainty. The Quarterly Journal of Experimental Psychology, 49, 546-571.

Hommel, B. (2009). Action control according to TEC (theory of event coding). Psychological Research, 73, 512-526.

Hommel, B., & Müsseler, J. (2006). Action-feature integration blinds to feature-overlapping perceptual events: Evidence from manual and vocal actions. The Quarterly Journal of Experimental Psychology, 59, 509-523.

Hommel, B., Müsseler, J., Aschersleben, G., & Prinz, W. (2001). The Theory of Event Coding (TEC): A new framework for perception and action planning. Behavioral and Brain Sciences, 24, 849-937.

James, W. (1890). The principles of psychology. Cambridge, MA: Harvard University Press.

Kirsch, W., & Kunde, W. (2012). Visual near space is scaled to parameters of current action plans. Journal of Experimental Psychology: Human Perception and Performance, 39, 1313-1325.

Kirsch, W., & Kunde, W. (2014). Impact of planned movement direction on judgments of visual locations. Psychological Research, 78, 705-720.

Koch, I., & Hoffmann, J. (2000). The role of stimulus-based and response-based spatial information in sequence learning. Journal of Experimental Psychology: Learning Memory and Cognition, 26, 863-882.

Kunde, W. (2001). Response-effect compatibility in manual choice reaction tasks. Journal of Experimental Psychology: Human Perception and Performance, 27, 387-394.

Linkenauger, S. A., Profitt, D. R., & Witt, J. K. (2011). Taking a hands-on approach: Apparent Grasping Ability Scales the per-ception of object size. Journal of Experimental Psychology: Human Perception and Performance, 37, 1432-1441.

Lotze, R. H. (1852). Medicinische Psychologie oder Physiologie der Seele [Medical psychology or physiology of the soul]. Berlin, Germany: Weidmann.

Morgado, N., Gentaz, E., Guinet, E., Osiurak, F., & Palluel-Germain, R. (2013). Within reach but not so reach-able: Obstacles matter in visual perception of distances. Psychonomic Bulletin and Review, 20, 462-467.

Lestage et al. 15

Morgado, N., & Palluel-Germain, R. (2015). How actions con-strain the visual perception of space. In Y. Coell & M. C. Fischer (Eds.), Foundations of embodied cognition: Perceptual and emotional embodiment (Vol. 1, pp. 167-180). Hove, UK: PsychologyPress.

Müsseler, J., & Hommel, B. (1997). Blindness to response-compatible stimuli. Journal of Experimental Psychology: Human Perception and Performance, 23, 861-872.

Noë, A. (2004). Action in perception. Cambridge: MIT Press.O’Regan, K. J., & Noë, A. (2001). A sensorimotor account of

vision and visual consciousness. Behavioral and Brain Sciences, 24, 939-973.

Osiurak, F., & Badets, A. (2014). Pliers, not fingers: Tool-action effect in a motor intention paradigm. Cognition, 130, 66-73.

Osiurak, F., & Badets, A. (2016). Tool use and affordance: Manipulation-based versus reasoning-based approaches. Psychological Review, 123, 534-568.

Osiurak, F., & Badets, A. (2017). Use of tools and misuse of embodied cognition. Psychological Review, 124, 361-368.

Pfister, R., Janczyk, M., Gressmann, M., Fournier, L., & Kunde, W. (2014). Good vibrations? Vibrotactile selfstimulation reveals anticipation of bodyrelated action effects in motor control. Experimental Brain Research, 232, 847-854.

Pfister, R., & Kunde, W. (2013). Dissecting the response in response-effect compatibility. Experimental Brain Research, 224, 647-655.

Rizzolatti, G., Riggio, L., & Sheliga, B. M. (1994). Space and selective attention. In C. Umilta, & M. Moscovitch (Eds.),

Attention and performance XV: Conscious and noncon-scious information processing (pp. 231-265). Cambridge: MIT Press.

Rusconi, E., Kwan, B., Giordano, B., Umiltà, C., & Butterworth, B. (2006). Spatial representation of pitch height: The SMARC effect. Cognition, 99, 113-129.

Shin, Y. K., Proctor, R. W., & Capaldi, E. J. (2010). A review of contemporary ideomotor theory. Psychological Bulletin, 136, 943-974.

Stefanucci J. K., & Geuss M. N. (2009). Big people, little world: The body influences size perception. Perception, 38, 1782-1795.

Stock, A., & Stock, C. (2004). A short history of ideo-motor action. Psychological Research, 68, 176-188.

Thébault, G., Michalland, A. H., Dérozier, V., Chabrier, S., & Brouillet, D. (2018). When the vibrations allow for antici-pating the force to be produced: An extend to Pfister et al. (2014). Experimental Brain Research, 236, 1219-1223. doi:10.1007/s00221-018-5190-x

van der Hoort, B., & Ehrsson, H. H. (2014). Body ownership affects visual perception of object size by rescaling the vis-ual representation of external space. Attention, Perception, & Psychophysics, 76, 1414-1428.

Veto, P., Einhäuser, W., & Troje, N. F. (2017). Biological motion distorts size perception. Scientific Reports, 7, 1-6.

Wykowska, A., Schubö, A., & Hommel, B. (2009). How you move is what you see: Action planning biases selection in visual search. Journal of Experimental Psychology: Human Perception and Performance, 35, 1755-1769.