Abstract

We used TMS to assess the causal roles of the lateral occipital (LO) and caudal intraparietal sulcus (cIPS) areas in the perceptual discrimination of object features. All participants underwent fMRI to localize these areas using a protocol in which they passively viewed images of objects that varied in both form and orientation. fMRI identified six significant brain regions: LO, cIPS, and the fusiform gyrus, bilaterally. In a separate experimental session, we applied TMS to LO or cIPS while the same participants performed match-to-sample form or orientation discrimination tasks. Compared with sham stimulation, TMS to either the left or right LO increased RTs for form but not orientation discrimination, supporting a critical role for LO in form processing for perception- and judgment-based tasks. In contrast, we did not observe any effects when we applied TMS to cIPS. Thus, despite the clear functional evidence of engagement for both LO and cIPS during the passive viewing of objects in the fMRI experiment, the TMS experiment revealed that cIPS is not critical for making perceptual judgments about their form or orientation.

INTRODUCTION

The ability to use vision to process object features, such as form and orientation, is important for guiding adaptive behavior. The consequences of being unable to do so are apparent in visual agnosia (for a review, see Farah, 2004). Patients with this condition are not blind. They can see objects but have difficulty either binding their features into a meaningful perception or associating a visual representation with a particular meaning, resulting in profound deficits in object recognition. A famous case of this condition is Oliver Sacks' patient Dr. P., who, upon mistaking his wife's head for his hat, reached out his hand in an attempt to grasp her head as though it were his hat (Sacks, 1985).

Patient and functional neuroimaging studies that have attempted to identify the important brain areas for the perceptual recognition of object form and orientation have yielded contradictory results. As we review below, some studies have demonstrated that the lateral occipital (LO) area, located in the ventral stream, is uniquely implicated in both object form and orientation recognition, whereas other studies have implicated the caudal intraparietal sulcus (cIPS) area, located in the dorsal stream, as a structure that is important for orientation recognition. Any conclusions of causality, however, are limited by the diffuse nature of the brain lesions in the patient studies and the correlational nature of functional neuroimaging. The goal of the present investigation was to use TMS to determine if form and orientation recognition are dependent on intact LO and cIPS processing.

Visual form agnosia is one of many types of visual agnosia. Patients with this agnosia manifest an inability to recognize objects due to problems in processing their shape (Milner et al., 1991). This inability is usually accompanied by impairments in processing the objects' orientation and width (Milner et al., 1991). Although rare, visual form agnosia is frequently associated in the literature with bilateral damage to the more posterior portions of the ventral stream (Farah, 2004; James, Culham, Humphrey, Milner, & Goodale, 2003). Interestingly, Goodale, Milner, Jakobson, and Carey (1991) describe a patient, D.F., who cannot perceive the form, orientation, and width of objects but who can nevertheless accurately guide her hand and finger movements in flight to grasp these same objects. At the time, structural imaging of D.F.'s brain revealed that she had bilateral ventral-stream damage in what we would now call the lateral occipital (LO) area, with relative sparing of her dorsal stream.

These data suggest that the neural substrates for the perceptual recognition of object features are distinct from those that guide the control of manual skills. On the basis of these data and a range of other neuropsychological, neurophysiological, and neuroanatomical studies, Goodale and Milner (1992) proposed that the dorsal and ventral streams serve unique roles. According to their proposal, the dorsal stream is critical for visually guiding the hand to an object, whereas the ventral stream is critical for perceiving and recognizing objects (Milner & Goodale, 2008; Goodale & Milner, 1992).

Although most types of visual agnosia are associated with damage to the ventral stream (for a review, see Farah, 2004; Goodale & Milner, 1992), some patient studies suggest that damage to the parietal lobes might also result in orientation agnosia (Harris, Harris, & Caine, 2001; Turnbull, Beschin, & Della Sala, 1997; Turnbull, Laws, & McCarthy, 1995). The patients described in these studies exhibit severe impairments in determining object orientation despite intact abilities in recognizing the form and identity of objects. The presence of this neurological dissociation suggests that object form and orientation might be mediated by independent neural mechanisms, with the latter requiring dorsal stream processes.

LO's involvement in form and orientation processing is supported by fMRI studies. Using fMRI adaptation, several studies have demonstrated that (1) the posterior subdivision of LO adapts to the form of objects when those objects are presented in the same orientation, and (2) the anterior subdivision of this same region adapts to the form of objects irrespective of orientation (e.g., Valyear, Culham, Sharif, Westwood, & Goodale, 2006; Kourtzi, Erb, Grodd, & Bulthoff, 2003; James, Humphrey, Gati, Menon, & Goodale, 2002; Vuilleumier, Henson, Driver, & Dolan, 2002; Grill-Spector et al., 1999). These findings suggest a role for LO in processing the form and orientation of objects.

Several fMRI adaptation studies, however, have shown that orientation processing is less specific to LO than form processing (e.g., Valyear et al., 2006; Kourtzi et al., 2003; James et al., 2002; Vuilleumier et al., 2002). The results of these studies indicate that the most caudal aspect of the IPS (cIPS) also adapts when different objects are presented in the same orientation. Furthermore, other fMRI studies show greater activation in the cIPS while participants actively attend to object orientation as compared with other object features, such as the color of objects (Shikata et al., 2001, 2003). These findings converge well with those from single-cell recording studies in monkeys, which show large concentrations of neurons in this area that selectively code for the orientation of stimuli (e.g., Tsutsui, Sakata, Naganuma, & Taira, 2002; Taira, Tsutsui, Jiang, Yara, & Sakata, 2000; Sakata et al., 1998).

Lesion studies experience the problem that brain damage is not always confined to a particular brain area unless that region is surgically excised. For example, D.F.'s visual form agnosia resulted from carbon monoxide poisoning, which can lead to diffuse damage in the brain. Recent structural analyses of D.F.'s brain using high-resolution MRI indicate greater damage to the dorsal stream than previously thought (Wood, Chouinard, Major, & Goodale, in press; Whitwell, Milner, & Goodale, 2014; Bridge et al., 2013). Likewise, the reports of orientation agnosia described earlier were conducted with patients who either had large lesions that extended beyond the parietal lobe (Turnbull et al., 1997), multiple infarcts throughout the brain as result of vascular disease (Turnbull et al., 1995), or Alzheimer disease with widespread neurodegeneration (Harris et al., 2001). Although brain damage in these patients was reported to be most prominent in the dorsal stream, their damage could have extended to the ventral stream.

Functional neuroimaging techniques, such as fMRI, also have important limitations. Data obtained from these methodologies represent correlations between hemodynamic responses and behavioral tasks and do not allow one to differentiate brain areas that are critical for task performance from those that are not. This limitation is particularly important to consider in parallel systems. For example, according to the two visual streams theory proposed by Goodale and Milner (1992), it could be the case that previous reports of orientation-selective fMRI activation in the cIPS might reflect processes that are critically important for visuomotor transformations but less important for perceptual decision-making.

In contrast to patient lesion and functional neuroimaging approaches, TMS provides an opportunity to infer causality in investigations of brain–behavior relationships in a highly controlled fashion (Chouinard & Paus, 2010). TMS perturbs neural activity by the delivery of brief magnetic pulses through the skull, which induce electrical currents in the underlying brain tissue. Experimentally manipulating brain function in this way and measuring its effect on task performance allow one to determine if a particular brain area is necessary. Thus, we used TMS to examine the necessity of LO and cIPS in object form and orientation perception by stimulating each of these regions while participants performed match-to-sample judgments on form or orientation.

We had two competing hypotheses. The first, based on dissociations described by some investigators in the agnosia literature (Harris et al., 2001; Turnbull et al., 1995, 1997), was that TMS to LO would affect the perceptual discrimination of form but not orientation, whereas TMS to cIPS would affect the perceptual discrimination of orientation but not form. The second, based on the two visual streams theory of Goodale and Milner (1992), was that TMS to LO would affect the perceptual discrimination of form and orientation, whereas TMS to cIPS would affect neither types of perceptual discrimination. The latter follows from the idea, central to Goodale and Milner's (1992) model, that visual processing carried out by the dorsal stream is not critical for the perceptual recognition of object features.

METHODS

Participants

Fourteen volunteers (nine men, mean age = 24.9 years, age range = 21–36 years) participated in the fMRI localization and TMS experiments. All participants self-reported to be right-handed and have normal or corrected-to-normal vision. None of the participants self-reported a history of neurological impairment. The research ethics board at the University of Western Ontario (London, Ontario, Canada) approved the procedures, and all participants provided informed written consent before their participation. We provided monetary compensation to the participants for their time and inconvenience.

Stimuli

Six grayscale novel objects composed of geons (e.g., cones, wedges, cylinders, and blocks; Biederman, 1987) were selected from Michael Tarr's stimulus data set (available at http://wiki.cnbc.cmu.edu/Novel_Objects). We carefully selected the stimuli so they would not carry any action affordances (i.e., qualities of an object that associate it with action; Vingerhoets, 2008). We used E-Prime 2.0 software (Psychology Software Tools, Pittsburgh, PA) to present the stimuli in both experiments.

Procedures for the fMRI Localizer Experiment

We collected data on a Siemens (Erlangen, Germany) 3-T Tim Trio MRI system using the posterior half of a 12-channel receiver-only head coil and a 4-channel flex coil over the forehead to recover anterior signal loss. We acquired an anatomical scan encompassing the whole brain. This was achieved by collecting 192 1-mm-thick slices using a 3-D T1-weighted acquisition sequence (inversion time = 900 msec, echo time = 3.43 msec, repetition time = 2300 msec, flip angle = 9°). The in-plane resolution of the anatomical scans was 256 × 240 pixels. BOLD data were collected using a T2*-weighted EPI sequence (echo time = 30.0 msec, repetition time = 3,000 msec, flip angle = 78°). The in-plane matrix size was 80 × 80 pixels with a field of view of 16.8 × 16.8 cm. Each volume was composed of 42 slices oriented parallel to the AC–PC plane (isotropic voxel size = 3.0 mm). There were no gaps between slices.

For localizing LO and cIPS, we designed our experiment such that we could subtract BOLD activation to scrambled images from intact versions of the same images presented at different orientations. The underlying rationale being that the effects from processing the low-level features of the images would be removed by the subtraction—leaving us with activation related only to the visual processing of form and orientation of the intact stimuli. In other words, the contrast would reveal form-selective areas, orientation-selective areas, and areas selective to both form and orientation. The intention of the fMRI experiment was not to dissociate form from orientation processing but rather to localize within each participant regions sensitive to one or both of these features for the purposes of perturbing them with TMS during match-to-sample judgment tasks on form and orientation.

To create the scrambled images, each image was cut into 196 pieces, which were randomly rearranged to form a new stimulus within the same space. In total, there were 72 stimuli (6 exemplars × 6 orientations for both intact and scrambled stimuli). During fMRI, the stimuli were projected onto a screen in the back of the MRI scanner. Participants viewed the stimuli from a mirror attached to the head coil. Participants were instructed to maintain fixation on a small dot in the center of the screen, which remained on-screen for each run. For a given block, 12 stimuli, either intact or scrambled, were presented. Each stimulus was presented in central vision for 750 msec at six different orientations, divisible by 60°, and subtended 4.5° of visual angle. Each block was followed by a 9-sec delay to allow the hemodynamic response to return to baseline before the start of the next block. The functional localizer was composed of two different runs; each run contained nine blocks of intact stimuli and nine blocks of scrambled stimuli. Each run consisted of 114 volumes corresponding to a duration of 5.7 min.

Analyses of the fMRI Localizer Experiment

We preprocessed and analyzed the functional data using SPM8 software (University College of London, United Kingdom). We discarded the first two volumes of each run because they were unlikely to represent steady state. We then carried out motion correction to align the data to the overall mean and applied a slice-time correction algorithm to resample every slice to match the first slice in each volume. Afterward, the data were coregistered to the anatomical MRI, upscaled to an isotropic voxel size of 1.0 mm, passed through a high-pass filter (cutoff = 128 sec), and passed through a Gaussian filter to achieve a final 6-mm FWHM of smoothing.

We then analyzed the data using a general linear model. Specifically, we modeled the time course of each block condition as a box function convolved with a standard double-gamma hemodynamic response function, estimated this model, and computed a t statistic for every voxel in the brain to assess whether or not there was greater BOLD activation to the intact compared with the scrambled conditions. The t statistical map for each individual (in native space) was superimposed over their anatomical MRI (also in native space) for the purposes of TMS localization (see Apparatus and Procedure for the TMS Experiment section). Thus, the fMRI results for each individual as opposed to the fMRI group results were used for TMS localization. We also spatially normalized the data into standardized space using the Montreal Neurological Institute (MNI) brain (McGill University, Montreal, Quebec, Canada) and carried out a random effects analysis to examine significant activation at the group level. To determine group-level significance, we performed a cluster analysis to correct for multiple comparisons in a whole-brain volume of 1000 cm3 (Worsley, Evans, Marrett, & Neelin, 1992). This yielded a threshold size of 258 mm3 (i.e., a threshold size of 258 voxels with an isotropic voxel size of 1.0 mm) for reporting a cluster of voxels with a p-uncorrected value of less than .001 as being significant at pcorr < .05 (Worsley et al., 1992).

Apparatus and Procedure for the TMS Experiment

We used a functional localization approach for determining the placement of the TMS coil. This approach consists of first localizing targets with fMRI in each individual and using an MRI-guided navigation system to guide the TMS coil over these sites defined at the participant level. The Brainsight neuronavigation system (Rogue Research, Montreal, Quebec, Canada) and a Polaris infrared camera (Northern Digital, Waterloo, Ontario, Canada) was used to place the TMS coil over LO and cIPS after a four-step procedure. First, we superimposed the activation map resulting from the fMRI localizer experiment over the participant's anatomical MRI scan in native space. Second, we marked each area of interest on the MRI while considering levels of peak activation and anatomical landmarks (i.e., LO had to be located in the LO sulcus, and cIPS had to be located in the most caudal and ventral aspect of the IPS). Third, we registered the participant's head to their MRI. For this registration, the participant wore a reference headband with small reflective balls so that the neuronavigation system could track the participant's head in space with the Polaris infrared camera. The experimenter, with a handheld pointer, indicated to the system where the participant's tip of the nose, bridge of the nose, and the notch above the tragus for each ear were located relative to the participant's reference headband, enabling us to then move and track the TMS coil in real time to target specific areas of the brain indicated on the MRI. Fourth, we placed the TMS coil over the target location. The TMS coil was manually held in place by the experimenter, who tracked its positioning in real time and made small adjustments as needed to ensure that stimulation was always on target. The participants placed their head on a chin rest during task performance and stimulation.

Biphasic pulses of TMS were delivered through a figure-eight coil connected to a Magstim rapid-rate stimulator (Magstim Company, Spring Gardens, UK; diameter of each coil = 7 cm). During task performance (see Procedures for the TMS experiment), TMS was delivered at an intensity of 120% of the resting motor threshold (rMT) for the right first dorsal interosseous muscle. This threshold was determined in the following manner. First, we placed disposable electrodes over the muscle (recording electrode), the interphalangeal joint (indifferent electrode), and the wrist (ground electrode). These electrodes were connected to the EMG module of the TMS system. Afterward, we placed the TMS coil over the left primary motor cortex by manually moving it over the scalp to the location where the strongest motor-evoked potential could be induced at a suprathreshold intensity. Once we found this location, we reduced the stimulation intensity incrementally until we could determine the lowest intensity required to reliably produce 5 of 10 motor-evoked potentials with peak-to-peak amplitudes of at least 50 μV. The resulting value was considered the rMT. The rMTs were considered only for tailoring stimulation intensities for each participant and were thus not systematically recorded.

Procedures for the TMS Experiment

In the match-to-sample visual discrimination tasks, participants indicated which of the two choice stimuli matched the sample stimulus they saw previously (Figure 1). There were two versions of the task. In the form discrimination task, the form of only one of the choice stimuli matched the form of the sample. The irrelevant stimulus feature, orientation, was the same for all stimuli in a trial but varied from trial to trial. In the orientation discrimination task, the orientation of only one of the choice stimuli matched the orientation of the sample stimulus. The irrelevant stimulus feature, form, was the same for all stimuli in a trial but varied from trial to trial.

Figure 1. 

Experimental paradigm. The figure illustrates the procedures used for this study. All participants underwent an fMRI experiment in which they passively viewed novel stimuli in their intact and scrambled configurations (A). The intact versions of the stimuli were presented in different orientations. Subtracting these two conditions enabled us to functionally define the locations for cIPS and LO in each of the two hemispheres on a participant-by-participant basis. After these areas were defined in a neuronavigation system, we then coregistered the participant's head with their MRIs by frameless strereotaxy and targeted each ROI marked on the participant's MRI. B shows a participant receiving TMS during task performance. C and D illustrate the events in a trial for both the form and orientation discrimination tasks, respectively. A trial began with the presentation of the sample image for 40 msec. This image was presented either to the left or right of fixation. A mask, consisting of a highly salient, colored Mondrian, was then presented for 500 msec. Two pulses of TMS separated by 100 msec were delivered at the onset of the presentation of the mask. After the presentation of the mask, two choice images were presented on the screen until a response was made by the participant indicating which of the two images they thought was previously shown as the sample. In the form discrimination task (C), the two choice images differed in form but not orientation. In the orientation discrimination task (D), the two choice images differed in orientation but not form.

Figure 1. 

Experimental paradigm. The figure illustrates the procedures used for this study. All participants underwent an fMRI experiment in which they passively viewed novel stimuli in their intact and scrambled configurations (A). The intact versions of the stimuli were presented in different orientations. Subtracting these two conditions enabled us to functionally define the locations for cIPS and LO in each of the two hemispheres on a participant-by-participant basis. After these areas were defined in a neuronavigation system, we then coregistered the participant's head with their MRIs by frameless strereotaxy and targeted each ROI marked on the participant's MRI. B shows a participant receiving TMS during task performance. C and D illustrate the events in a trial for both the form and orientation discrimination tasks, respectively. A trial began with the presentation of the sample image for 40 msec. This image was presented either to the left or right of fixation. A mask, consisting of a highly salient, colored Mondrian, was then presented for 500 msec. Two pulses of TMS separated by 100 msec were delivered at the onset of the presentation of the mask. After the presentation of the mask, two choice images were presented on the screen until a response was made by the participant indicating which of the two images they thought was previously shown as the sample. In the form discrimination task (C), the two choice images differed in form but not orientation. In the orientation discrimination task (D), the two choice images differed in orientation but not form.

For both tasks, a trial began with the brief presentation of the sample image for 40 msec. Half of the sample stimuli were presented in the right visual field, whereas the other half were presented in the left visual field. A mask was then presented for 500 msec. Specifically, we opted to present a highly salient, colored Mondrian as a mask to help counter the possibility of a TMS-induced reduction in mask effectiveness. Two pulses were applied at 100-msec intervals (10 Hz) immediately after the onset of the mask presentation—leaving 400 msec of mask presentation without TMS. After the presentation of the mask, the two choice images appeared and remained onscreen until the participants made a response on a PST Serial Response Box (Psychology Software Tools). Specifically, participants pressed buttons “1” and “2” to select the top and bottom choice images, respectively. E-Prime randomly generated the order of presentation.

The sample and choice stimuli subtended a visual angle of 4.5° with the midpoint of these images appearing at 6.5° eccentricity from central fixation. The length of the intertrial period varied randomly among 8.0, 8.5, and 9.0 sec to prevent the participant from being able to anticipate the sample stimulus and to allow enough time to separate each trial, as recommended by published TMS safety guidelines (Wassermann, 1998). Participants were instructed to maintain their gaze on the fixation cross, which remained present throughout each block. In the orientation discrimination task, the two choice stimuli differed in their orientation by 32°.

Each participant completed two practice blocks of 40 trials (one per discrimination task) and 10 experimental blocks of 40 trials (five per discrimination task). Thus, there were 20 trials per condition if one considers that the stimulus was presented in the left visual field for half of the trials in a given block and the right visual field in the remaining half. The participants performed the practice blocks at the beginning of the TMS session so they could become acquainted with the tasks. The participants then completed two experimental blocks (one per discrimination task) with TMS applied to one of five possible sites (i.e., left LO, right LO, left cIPS, right cIPS, or sham). This was repeated for each of the remaining sites. We counterbalanced the order of stimulation sites across participants. Blocks alternated between form and orientation discrimination tasks, starting with form in half of the participants and orientation in the other half of the participants. For LO stimulation, we held the TMS coil tangentially to the scalp with the handle parallel to the vertical plane. For cIPS stimulation, we held the TMS coil tangentially to the scalp with the handle angled 45° upward from the horizontal plane. We applied sham stimulation with the edge of the coil resting along the longitudinal fissure with the coils orientated away from the scalp such that the induced magnetic field was directed away from the head.

Monitoring the Peripheral Effects of TMS

Although we did not monitor central fixation using an eye tracker, we did monitor eye twitching by having one experimenter look at the participant's eyes while a different experimenter applied TMS. We adjusted the orientation of the TMS coil to help minimize eye twitching in participants who demonstrated it. As noted above,we delivered two pulses of TMS after the sample image disappeared, leaving at least 250 msec for any eye twitching to recover before we presented the participant with the two choice images. Thus, it seems unlikely that eye twitching could have affected the retinal presentation of either the sample or choice stimuli. In addition, we asked the participant how the TMS felt after each block of stimulation.

Analyses of the TMS Experiment

We carried out most of the statistical analyses using the Statistical Package for the Social Sciences (SPSS; IBM Corporation, Armonk, NY). Outliers were removed by computing RT z scores for each condition at the participant level and removing all trials exceeding a score below −3.0 or above +3.0. After the removal of the outliers, the mean RT (milliseconds) and accuracy (% correct) were calculated for each participant. We then performed a repeated-measures ANOVA with Stimulation site (left cIPS vs. right cIPS vs. left LO vs. right LO vs. sham) and Visual field (left visual field vs. right visual field) as factors. Tukey's honest significant difference tests, which correct for multiple comparisons, were carried out after using procedures described elsewhere (Tukey, 1949). Unless specified otherwise, all reported p values were corrected for multiple comparisons, and statistical significance was evaluated in reference to an alpha level of .05 for two-tailed tests. For reporting effect sizes, we calculated Cohen's f from the partial eta-squared (η2) values obtained from the SPSS ANOVA output table (calculated as η2/(1 − η2)) and Cohen's d for pairwise comparisons as the difference between the two means divided by their pooled standard deviation (Cohen, 1988).

Depiction of the Sites of Stimulation

For illustrative purposes, we transformed the Brainsight coordinates for the locations of the stimulation sites into MNI space and then manually mapped these sites onto a cortical reconstruction of each participant's hemisphere (Figure 2). These reconstructions were created using Freesurfer's recon-all function (Martinos Center for Biomedical Imaging, Harvard University, Cambridge, MA). The mean ± SD MNI coordinates for each stimulation site, as defined in volumetric space in Brainsight as opposed to pial space in Freesurfer, are as follows: x = −23.3 ± 5.97, y = −78.9 ± 7.10, z = 32.8 ± 9.63 for the left cIPS; x = −27.2 ± 6.71, y = −78.3 ± 5.47, z = 35.6 ± 7.62 for the right cIPS; x = −40.4 ± 6.16, y = −82.3 ± 7.94, z = −2.0 ± 6.97 for the left LO; and x = 47.4 ± 4.10, y = −75.9 ± 5.55, z = −4.1 ± 6.26 for the right LO.

Figure 2. 

Stimulation sites. For illustrative purposes, the stimulation sites (blue: cIPS, LO: orange) for each participant were manually marked on 3-D cortical reconstructions created in Freesurfer based on the stimulation coordinates defined in volumetric space. The reconstructions presented in the figure show the lateral convexity of the two hemispheres being viewed from the back. The sites were determined from each participant's fMRI localizer contrast while also considering anatomical landmarks. cIPS was defined as the peak activation inside or near the most caudal aspect of the IPS, whereas LO was defined as the peak activation inside or near the LO sulcus.

Figure 2. 

Stimulation sites. For illustrative purposes, the stimulation sites (blue: cIPS, LO: orange) for each participant were manually marked on 3-D cortical reconstructions created in Freesurfer based on the stimulation coordinates defined in volumetric space. The reconstructions presented in the figure show the lateral convexity of the two hemispheres being viewed from the back. The sites were determined from each participant's fMRI localizer contrast while also considering anatomical landmarks. cIPS was defined as the peak activation inside or near the most caudal aspect of the IPS, whereas LO was defined as the peak activation inside or near the LO sulcus.

RESULTS

All participants tolerated the procedures well. Eye twitching to TMS was detected in some participants during LO stimulation, which was reduced by adjusting the orientation of the TMS coil. After these adjustments, eye twitching, when present, was mild and unlikely to confound interpretation of the data. We note further that all participants reported LO to be the least comfortable stimulation site but none reported minding the stimulation applied to this site.

fMRI Localization Experiment

A random effects analysis at the group level revealed four clusters of significant activation, each corresponding to our predicted areas of interest in the two hemispheres (Figure 3, Table 1). The four clusters consisted of the LO complex (including peak foci of activation in LO and the fusiform gyrus) in both hemispheres and the cIPS in both hemispheres (Figure 3, Table 1). No other clusters were significant. Euclidean distances between the average sites that were actually stimulated and the group foci of activation were as follows: 8.0 mm for the left cIPS, 5.2 mm for the right cIPS, 2.5 mm for the left LO, and 6.4 mm for the right LO.

Figure 3. 

Results from the fMRI localizer experiment. The figure shows clusters of significant activation for all 14 participants (pcorr < .05). Namely, the statistical map was set at a cluster threshold for puncorr < .001 and an extent greater than 258 voxels (or 258 mm3; see Methods for more details). This analysis revealed four significant clusters: (1) the left cIPS, (2) the right cIPS, (3) the left LO along with the left fusiform (Fus) as part of the same cluster, and (4) the right LO along with the right fusiform (Fus) as part of the same cluster. Images are presented in neurological convention (i.e., left is left; right is right).

Figure 3. 

Results from the fMRI localizer experiment. The figure shows clusters of significant activation for all 14 participants (pcorr < .05). Namely, the statistical map was set at a cluster threshold for puncorr < .001 and an extent greater than 258 voxels (or 258 mm3; see Methods for more details). This analysis revealed four significant clusters: (1) the left cIPS, (2) the right cIPS, (3) the left LO along with the left fusiform (Fus) as part of the same cluster, and (4) the right LO along with the right fusiform (Fus) as part of the same cluster. Images are presented in neurological convention (i.e., left is left; right is right).

Table 1. 

Results from the fMRI Localizer Experiment: Subtraction between the Intact and Scrambled Objects Presented at Different Orientations

Brain AreaAnatomical LandmarkxyzT(13)Cluster Size, p < .001 (mm3)
Right LO LO sulcus 46 −77 13.6 4930 
Right Fus Fusiform gyrus 45 −59 −12 6.1 
Left LO LO sulcus −42 −84 −1 9.6 4998 
Left Fus Fusiform gyrus −39 −59 −15 6.2 
Right cIPS Caudal portion of the intraparietal sulcus 24 −75 38 5.0 502 
Left cIPS Caudal portion of the intraparietal sulcus −23 −77 25 5.4 314 
Brain AreaAnatomical LandmarkxyzT(13)Cluster Size, p < .001 (mm3)
Right LO LO sulcus 46 −77 13.6 4930 
Right Fus Fusiform gyrus 45 −59 −12 6.1 
Left LO LO sulcus −42 −84 −1 9.6 4998 
Left Fus Fusiform gyrus −39 −59 −15 6.2 
Right cIPS Caudal portion of the intraparietal sulcus 24 −75 38 5.0 502 
Left cIPS Caudal portion of the intraparietal sulcus −23 −77 25 5.4 314 

All coordinates are in MNI space.

TMS Experiment: Form Discrimination Task

TMS slowed RTs during form discrimination when it was applied to either the left or right LO (Figure 4). Specifically, there was a main effect of Stimulation site (F(4, 52) = 4.61, p = .003, f = 0.596), and pairwise comparisons revealed that stimulation to each of these two sites disrupted RTs relative to sham (left LO vs. sham: p = .033, right LO vs. sham: p = .003; Figure 4A). Conversely, pairwise comparisons did not reveal any differences in RT when stimulation was applied to cIPS relative to sham (left cIPS vs. sham: p = .758, right cIPS vs. sham: p = .315; Figure 4A). No differences were found between LO and cIPS stimulation: all ps > .068). Table 2A provides descriptive statistics and Cohen's d for these different pairwise comparisons. Neither the main effect of Visual field (F(1, 13) = 2.15, p = .166, f = 0.407; Figure 4B) nor its interaction with Stimulation site (F(4, 52) = 0.99, p = .423, f = 0.276) was significant.

Figure 4. 

TMS experiment: form discrimination task. A shows how the RTs were slower when TMS was applied over both the left and right LO compared with sham. B shows how the RTs did not differ when the sample stimuli were presented in the left relative to the right visual field. C shows how TMS did not affect accuracy. D shows how the participants were more accurate in discriminating the form of objects when they were presented in the right relative to the left visual field. Interactions between Stimulation site and Visual field were not significant. Error bars represent SEMs. Asterisks (*) denote significant differences after corrections were applied for multiple comparisons (pcorr < .05).

Figure 4. 

TMS experiment: form discrimination task. A shows how the RTs were slower when TMS was applied over both the left and right LO compared with sham. B shows how the RTs did not differ when the sample stimuli were presented in the left relative to the right visual field. C shows how TMS did not affect accuracy. D shows how the participants were more accurate in discriminating the form of objects when they were presented in the right relative to the left visual field. Interactions between Stimulation site and Visual field were not significant. Error bars represent SEMs. Asterisks (*) denote significant differences after corrections were applied for multiple comparisons (pcorr < .05).

Table 2. 

Descriptive Statistics and Effect Sizes for the Pairwise Comparisons between Stimulation Site Conditions in the Form Discrimination Task

Condition (1)M (1)SD (1)Condition (2)Cohen's d
A. Pairwise comparisons between stimulation site conditions (RTs in milliseconds) 
Left cIPS 710.39 228.28 Right cIPS 0.116 
Left LO 0.257 
Right LO 0.388 
Sham 0.220 
Right cIPS 739.55 279.68 Left LO 0.140 
Right LO 0.260 
Sham 0.316 
Left LO 781.45 326.52 Right LO 0.110 
Sham 0.441* 
Right LO 663.66 320.95 Sham 0.579* 
 
B. Pairwise comparisons between stimulation site conditions (accuracy in percentages) 
Left cIPS 90.26 8.48 Right cIPS 0.050 
Left LO 0.064 
Right LO 0.131 
Sham 0.027 
Right cIPS 89.72 13.11 Left LO 0.013 
Right LO 0.051 
Sham 0.023 
Left LO 89.55 13.61 Right LO 0.034 
Sham 0.036 
Right LO 89.99 8.67 Sham 0.083 
Condition (1)M (1)SD (1)Condition (2)Cohen's d
A. Pairwise comparisons between stimulation site conditions (RTs in milliseconds) 
Left cIPS 710.39 228.28 Right cIPS 0.116 
Left LO 0.257 
Right LO 0.388 
Sham 0.220 
Right cIPS 739.55 279.68 Left LO 0.140 
Right LO 0.260 
Sham 0.316 
Left LO 781.45 326.52 Right LO 0.110 
Sham 0.441* 
Right LO 663.66 320.95 Sham 0.579* 
 
B. Pairwise comparisons between stimulation site conditions (accuracy in percentages) 
Left cIPS 90.26 8.48 Right cIPS 0.050 
Left LO 0.064 
Right LO 0.131 
Sham 0.027 
Right cIPS 89.72 13.11 Left LO 0.013 
Right LO 0.051 
Sham 0.023 
Left LO 89.55 13.61 Right LO 0.034 
Sham 0.036 
Right LO 89.99 8.67 Sham 0.083 

Asterisks (*) denote significant pairwise comparisons (pcorr < .05).

Although TMS to LO increased RTs, it had no effect on accuracy. Neither the main effect of Stimulation site (F(4, 52) = 0.10, p = .983, f = 0.084; Figure 4C) nor its interaction with Visual field (F(4, 52) = 1.01, p = .413, f = 0.279) was significant. Nonetheless, participants were better at discriminating form when the sample image was presented in the right as compared with the left visual field (F(1, 13) = 4.97, p = .048, f = 0.605; Figure 4D). Table 2B provides descriptive statistics and Cohen's d for pairwise comparisons for accuracy between stimulation sites.

TMS Experiment: Orientation Discrimination Task

TMS did not affect RTs or accuracy on the orientation discrimination task (Figure 5). In terms of RTs, there was neither a main effect of Stimulation site (F(4, 52) = 1.22, p = .313, f = 0.307; Figure 5A) and Visual field (F(1, 13) = 0.88, p = .367, f = 0.259; Figure 4B) nor an interaction between these two factors (F(4, 52) = 0.77, p = .548, f = 0.244). Paralleling the RT analysis, there was neither a main effect of Stimulation site (F(4, 52) = 0.89, p = .475, f = 0.264; Figure 5C) nor a main effect of Visual field (F(1, 13) = 0.18, p = .678, f = 0.123; Figure 5D) on accuracy. Similarly, Stimulation site and Visual field did not interact (F(4, 52) = 0.41, p = .799, f = 0.179). Table 3 provides descriptive statistics and Cohen's d for the different conditions presented in Figure 5.

Figure 5. 

TMS experiment: orientation discrimination task. The figure illustrates the lack of main effects for Stimulation site (A: RT, C: accuracy) and Visual field (B: RT, D: accuracy) during the orientation discrimination task. Interactions between these two factors were also not significant. Error bars represent SEMs.

Figure 5. 

TMS experiment: orientation discrimination task. The figure illustrates the lack of main effects for Stimulation site (A: RT, C: accuracy) and Visual field (B: RT, D: accuracy) during the orientation discrimination task. Interactions between these two factors were also not significant. Error bars represent SEMs.

Table 3. 

Descriptive Statistics and Effect Sizes for the Pairwise Comparisons between Stimulation Site Conditions in the Orientation Discrimination Task

Condition (1)M (1)SD (1)Condition (2)Cohen's d
A. Pairwise comparisons between stimulation site conditions (RTs in milliseconds) 
Left cIPS 843.35 251.37 Right cIPS 0.151 
Left LO 0.140 
Right LO 0.292 
Sham 0.084 
Right cIPS 882.80 280.42 Left LO 0.003 
Right LO 0.126 
Sham 0.218 
Left LO 883.69 328.31 Right LO 0.111 
Sham 0.205 
Right LO 820.41 255.05 Sham 0.349 
 
B. Pairwise comparisons between stimulation site conditions (accuracy in percentages) 
Left cIPS 87.61 9.14 Right cIPS 0.105 
Left LO 0.296 
Right LO 0.206 
Sham 0.006 
Right cIPS 86.57 10.93 Left LO 0.199 
Right LO 0.088 
Sham 0.090 
Left LO 84.03 14.80 Right LO 0.129 
Sham 0.256 
Right LO 87.69 10.28 Sham 0.167 
Condition (1)M (1)SD (1)Condition (2)Cohen's d
A. Pairwise comparisons between stimulation site conditions (RTs in milliseconds) 
Left cIPS 843.35 251.37 Right cIPS 0.151 
Left LO 0.140 
Right LO 0.292 
Sham 0.084 
Right cIPS 882.80 280.42 Left LO 0.003 
Right LO 0.126 
Sham 0.218 
Left LO 883.69 328.31 Right LO 0.111 
Sham 0.205 
Right LO 820.41 255.05 Sham 0.349 
 
B. Pairwise comparisons between stimulation site conditions (accuracy in percentages) 
Left cIPS 87.61 9.14 Right cIPS 0.105 
Left LO 0.296 
Right LO 0.206 
Sham 0.006 
Right cIPS 86.57 10.93 Left LO 0.199 
Right LO 0.088 
Sham 0.090 
Left LO 84.03 14.80 Right LO 0.129 
Sham 0.256 
Right LO 87.69 10.28 Sham 0.167 

TMS Experiment: Differences between Discrimination Tasks

We performed post hoc ANOVA to compare directly the effects of LO stimulation between the two discrimination tasks. Overall, these analyses revealed that the tasks differed in difficulty and that the effects of TMS did not differ significantly between them. For RTs, ANOVA with Task (form vs. orientation) × Site (left LO vs. right LO vs. sham) × VF (left vs. right) as within-participant factors revealed a main effect of Task (F(1, 13) = 25.17, p < .0001, f = 1.933) driven by faster RTs in the form discrimination task. Task did not interact with any of the other factors (all ps > .299). For accuracy, a similar ANOVA revealed a main effect of Task (F(1, 13) = 6.46, p = .025, f = 0.497) driven by more accurate performance in the form discrimination task.

DISCUSSION

Although the fMRI experiment revealed clear functional evidence that LO and cIPS are both engaged when people view visual objects varying in both form and orientation, the TMS experiment supported a critical role for form processing only for LO. Specifically, TMS to either the left or right LO increased RTs for form discrimination only, whereas TMS to either the left or right cIPS had no discernible effects on performance. However, as we will argue, there is some evidence from our data, and from work performed by others, that a subdivision of LO adjacent to the one we stimulated might fulfil a role in orientation processing. Together, these findings favored our second hypothesis, which was based on Goodale and Milner's (1992) two-stream hypothesis. Namely, perturbing activity in the ventral stream made a perceptual task more difficult, whereas perturbing activity in the dorsal stream did not. In the ensuing discussion, we will go over methodological issues before moving on to discussions on the effects of visual field presentation and stimulating each of our ROIs.

Methodological Considerations

Different orientation increments were presented in the fMRI (60°) and TMS (32°) experiments. This should be of no concern given the spatial resolution of our procedures. Hubel and Wiesel (1968) have shown that the representation for different orientations in the cortex is differentiated at the level of cortical columns that are no more than 500  μm in width (Mountcastle, 1957), which cannot be spatially differentiated using our fMRI procedures with a sampling resolution of 3.0 mm nor with our TMS figure-of-eight coil (diameter = 7 cm), which stimulates an estimated volume of 20 × 20 × 10 mm (Wassermann et al., 1996; Cohen et al., 1990; Maccabee et al., 1990).

TMS affected RT but not accuracy—ruling out speed–accuracy trade-offs. The absence of an effect of TMS on accuracy might come as a surprise in light of the virtual lesion metaphor sometimes used to describe TMS effects (e.g., Pascual-Leone, Walsh, & Rothwell, 2000). However, this metaphor should not be taken too literally. TMS can have facilitatory as well as disruptive effects depending on the task, stimulation parameters, and stimulation site (for a review, see Sandrini, Umilta, & Rusconi, 2011). Disruptive effects, when present, are rarely as complete or definite as those seen in patients with brain lesions. For these reasons, perturbation (Chouinard & Paus, 2010; Paus, 2005) might be a better descriptor of the effects of TMS than virtual lesion. In this study, the prolonged RTs could reflect a decrease in processing efficiency in the absence of a speed–accuracy trade-off. Indeed, another TMS study, applying stimulation to the extrastriate body area during the visual discrimination of body parts in a similar two-choice matching-to-sample paradigm, also report prolonged RTs without changes in accuracy (Urgesi, Candidi, Ionta, & Aglioti, 2007).

In addition, it seems unlikely that the increased discomfort and eye blinking during LO stimulation could explain our reported dissociation between the form and orientation discriminations tasks. Although a pilot experiment ensured roughly equivalent accuracy between the form and orientation discrimination tasks, RTs were still greater for the latter. Namely, TMS to LO disrupted performance on the form discrimination task, although the mean RT with sham TMS on this task was only 663.66 (SD = 203.43) msec. In contrast, TMS to LO did not disrupt performance on the orientation discrimination task, although the mean RT with sham TMS on this task was 820.41 (SD = 300.86) msec (Figures 4 and 5). It seems reasonable to assume that difficulty (as measured by RT) would be positively associated with the likelihood of TMS inducing a disruption in performance. In other words, it would have seemed more likely that performance on the orientation discrimination task would have been disrupted by TMS had any of our target sites been critically implicated in the perceptual processing of object orientation.

Last, in designing this study, we opted not to perform the same paradigm during the fMRI experiment as the TMS experiment for the following reasons. First, the purpose of this study was not to compare outcomes between two different methodologies but rather to use TMS to determine whether two functionally defined areas, cIPS and LO, play a necessary role in form and orientation discrimination. Second, our localizer task was a more cost-effective and time-efficient way to achieve this purpose. Regardless, we do not believe that a match-to-sample paradigm for the fMRI experiment would have made a difference for the purposes of TMS localization anyway. The location of activation for cIPS and LO would have likely remained the same given that the site of functionally defined higher-order visual areas tends to be context invariant (for a review, see Saxe, Brett, & Kanwisher, 2006) and that the same exemplars of objects were presented in both experiments.

The Right Visual Field Advantage for Discriminating Object Form

Presenting stimuli in different hemifields provides the opportunity to examine how different classes of stimuli might preferentially be processed by the left or right hemisphere. The tachistoscopic presentation of visual stimuli in the different hemifields has a long history as a method for examining hemispheric specialization (Sergent, 1983). For example, numerous studies using similar paradigms have shown a right visual field advantage for the recognition of printed words, which is consistent with the notion that the left but not the right hemisphere processes language in most individuals (Hunter & Brysbaert, 2008; Finkbeiner, Almeida, & Caramazza, 2006; Knecht et al., 2000; Bub & Lewine, 1988; Chiarello, Nuding, & Pollock, 1988). We did not have any a priori predictions as to whether there would be a right or left visual field advantage for our tasks. However, as it turns out, our participants had a right visual field advantage for discriminating the form of objects. This was demonstrated by greater accuracy for the sample images presented in the right relative to the left hemifield, irrespective of the hemisphere that was stimulated. However, we do not think that this finding necessarily reflects a left hemisphere specialization for object form discrimination. Rather, it is possible that the participants learned to recognize and even attribute lexical labels to each of the six different form exemplars in this study—introducing a language component to the task, which was not present in the orientation task. Further investigation is required to determine if this was the case.

LO Stimulation Disrupts Visual Form Processing

TMS to either the left or right LO increased form discrimination RTs, supporting a critical role for LO in form processing. Given that visual form agnosia arising from LO damage is usually accompanied by difficulties in perceiving the orientation of objects, one might also expect to find TMS-induced disruption in orientation discrimination as well. This was not the case. Stimulating LO in either hemisphere did not affect performance on the orientation discrimination task—although there was nevertheless a medium-to-large effect size (f = 0.309; Cohen, 1988) for this null finding.

A viable, although speculative, explanation for all of this is that we applied stimulation more directly to the subdivision of LO that mediates form processing than to the adjacent subdivision that mediates orientation processing. Simply, we may have only indirectly affected the adjacent orientation area as a result of a spread of current. Indeed, LO is recognized as encapsulating at least two major subdivisions: a posterior (LO1) and an anterior (LO2) subdivision (Larsson & Heeger, 2006). Each of these subdivisions contains their own topographic retinal representation of the contralateral hemifield (Larsson & Heeger, 2006) and adapts selectively to different types of object features (e.g., Larsson, Landy, & Heeger, 2006; Grill-Spector et al., 1999). LO1 adapts to the form of objects but only when objects are presented in the same viewpoint (e.g., Kourtzi et al., 2003; Vuilleumier et al., 2002; Grill-Spector et al., 1999). Conversely, LO2 still adapts to object form when objects are presented in different viewpoints (e.g., Valyear et al., 2006; Kourtzi et al., 2003; James et al., 2002; Vuilleumier et al., 2002; Grill-Spector et al., 1999). In other words, “retinal” and “perceived” form seems to be processed by LO1 and LO2, respectively.

In our experiments, we did not differentiate between subdivisions of LO. This is because we did not think this was possible using our figure-of-eight coil (diameter = 7.0 cm), which stimulates an estimated volume of 20 × 20 × 10 mm (Wassermann et al., 1996; Cohen et al., 1990; Maccabee et al., 1990). Instead, we chose to target our stimulation at the strongest foci of activation within the LO region, which tended to be in the most anterior subdivision of this region. However, Silson et al. (2013) have shown dissociable TMS-induced effects from stimulating LO1 and LO2 separately (range in the reported distance between the two sites across individuals was 6–15 mm) using a smaller and conceivably more focal figure-of-eight coil (diameter = 5 cm). Specifically, they showed that TMS to LO1 disrupted orientation but not shape discrimination, whereas TMS to LO2 disrupted shape but not orientation discrimination. The authors concluded from these results that orientation and shape are processed independently of one another in different subdivisions of LO. Based on of the reported coordinates of where they targeted, we estimate that we stimulated on average 7 and 18 mm anterior to their LO2 and LO1, respectively. Thus, orientation discrimination is likely causally related to LO1, a ventral stream processing region, but this region was not specifically targeted by our TMS.

Two other TMS studies, one from our laboratory and one from another, have disrupted form processing by stimulating the more anterior regions of LO (Chouinard, Whitwell, & Goodale, 2009; Pitcher, Charles, Devlin, Walsh, & Duchaine, 2009). In our study, we demonstrated how stimulating this region of LO in either hemisphere (MNI coordinates for left LO: x = −52, y = −62, z = −12; MNI coordinates for right LO: x = 40, y = −76, z = −16) disrupted the naming of visually presented objects—presumably by having disrupted form processing, given that a control experiment demonstrated that the same stimulation did not affect people's ability to name the color of these same objects (Chouinard et al., 2009). In the other study, Pitcher et al. (2009) demonstrated how stimulating the more anterior portion of LO in the right hemisphere (MNI coordinates: x = 44, y = −75, z = −6) disrupted a task in which participants indicated whether a probe consisting of stimuli obtained from the same database as the stimuli used in the present investigation was the same or different as a sample presented earlier.

The Lack of Task Disruption from cIPS Stimulation

Despite the clear fMRI evidence that LO and cIPS were engaged as the participants passively viewed objects varying in both form and orientation, the TMS results showed that the cIPS was not necessary for making form or orientation judgments about these same objects. It is unlikely that this null result reflected a lack of power for two reasons. First, the same stimulation applied to LO disrupted form discrimination at that site. Second, our clusters of activation in cIPS were smaller in spatial extent than those in LO. Thus, cIPS afforded us greater stimulation precision, minimizing error in our findings from functional–anatomical variation in this site.

In addition, we do not think that the lack of an effect was due to a timing issue. One could wonder whether the TMS to cIPS was delivered too early and that this area might process information after 100 msec, which was the time relative to onset that the last TMS pulse was delivered. We do not think this is a possibility for a number of reasons.

First, high-density electrophysiological mapping in humans, by means of either electroencephalography and magnetoencephalography, has shown that visual information reaches V1 25–50 msec after stimulus onset (Bar et al., 2006; Inui & Kakigi, 2006; Foxe & Simpson, 2002; for a review, see Laycock, Crewther, & Crewther, 2007) and the extrastriate visual areas in both the dorsal and ventral streams after 40–80 msec (Inui & Kakigi, 2006; Foxe & Simpson, 2002). Second, these same studies confirm that visual information arrives in the dorsal stream sooner than it does in the ventral stream (Bar et al., 2006; Inui & Kakigi, 2006; Foxe & Simpson, 2002). This finding is also well documented in the monkey, in which there are more inputs from the magnocellular system in the dorsal than the ventral stream (for a review, see Laycock et al., 2007). Therefore, had our stimulation been presented too early, it would have been much more of an issue for a ventral stream area, such as LO, than it would have been for a dorsal stream area, such as cIPS.

Third, using a design somewhat similar to the present investigation, Ellison and Cowey (2006) delivered TMS to both LO and a dorsal stream area, which they refer to as the posterior parietal cortex (PPC), to determine if each of these areas was necessary for shape, position, and color discrimination. However, they used a longer train of stimulation than we did, delivering their pulses at 10 Hz for 500 msec. Even with these longer trains of stimulation, they could not demonstrate TMS-induced effects for shape discrimination from stimulating PPC, although this did disrupt position discrimination. Similar to our study, Ellison and Cowey (2006) demonstrate that TMS applied to LO disrupts shape processing. Color processing was not affected by TMS applied to either LO or PPC. For all these reasons then, we can rule out the possibility that we stimulated cIPS too early to disrupt visual processing. It is possible, of course, that processes related to the cognitive manipulation of this information in the parietal lobe could still occur at a later time point (see below).

According to Goodale and Milner's (1992) two-stream theory, processing in dorsal stream areas of the parietal cortex should be critical for the use of vision in the online control of hand movements toward objects. In support of this notion, it has been demonstrated that the strength in functional coupling between cIPS and the anterior IPS, an area known to be important for shaping the fingertips on objects, increases as participants are required to either imagine or pantomime grasping movements to objects presented visually in different orientations (Shikata et al., 2003). Future experiments applying cIPS while participants guide their hand towards elongated objects that are visually presented in different orientations would be required to confirm this possibility.

It should also be noted that the cIPS we stimulated constitutes a different area of the parietal cortex than the one under investigation in earlier TMS studies by Harris and colleagues (Harris, Benito, Ruzzoli, & Miniussi, 2008; Harris & Miniussi, 2003). These studies show how TMS to a more superior part of the parietal cortex in the right hemisphere can interfere with orientation discrimination in the context of mental rotation (Harris & Miniussi, 2003; MNI coordinates: x = 34, y = −64, z = 50, representing a 20.8-mm difference with our average stimulation site for the right cIPS) and a task requiring participants to indicate whether or not a test arrow matched the long axis of an object in a previously shown photograph (Harris et al., 2008; MNI coordinates: x = 43, y = −63, z = 41, representing a 22.4-mm difference with our average stimulation site for the right cIPS).

Thus, in opposition to our findings, it would seem that Harris and colleagues demonstrated a role for dorsal stream regions in orientation processing. However, we believe that their studies likely required cognitive processes above and beyond those required to make simple visual orientation discriminations. In support of this interpretation, a comparison of their data during sham conditions to our own reveals longer RTs and lower accuracy. The increased task difficulty may reflect additional processing by the parietal lobe related to spatial working memory (Jonides et al., 1993), attention (Corbetta & Shulman, 2002), and/or the analysis of complex spatial relationships (Jordan, Heinze, Lutz, Kanowski, & Jancke, 2001). Therefore, it could be the case that the TMS-induced disruption reported in the studies by Harris and colleagues (Harris et al., 2008; Harris & Miniussi, 2003) had more to do with a disruption in higher-level cognitive operations than a disruption in the visual analysis of orientation. After all, damage to the right parietal lobe can lead to constructional apraxia (Russell et al., 2010), neglect (Karnath, 1997), as well as impairments in mental rotation (Ditunno & Mann, 1990).

Closing Remarks

The findings of this study provide three important take-home messages. The first is a cautionary tale: fMRI activation may not always reflect processes that are necessary for the successful completion of a particular task. Although cIPS was clearly engaged during the passive viewing of objects that varied in form and orientation, it did not play a critical role in the perceptual discrimination of either the form or orientation of objects. Second, these findings highlight a necessity to couple powerful imaging techniques such as fMRI with other investigative tools that are capable of testing causal statements relating brain and behavior. Finally, the findings from this study provide further support for the notion that the ventral but not the dorsal stream network plays a causal role in vision for perception.

Despite this, it is important to note that the results from this study do not imply a total lack of involvement of the dorsal stream in perception. It could be that the dorsal stream aids recognition—particularly for objects that can be manipulated—which is perhaps why areas in this stream tend to show greater activation for the viewing of manipulable versus nonmanipulable objects (Chouinard & Goodale, 2010). In light of the Goodale and Milner (1992) theory, one outstanding question concerns whether or not the cIPS is necessary for the visual guidance of the hand. Other TMS studies have previously confirmed a causal role of the anterior IPS in the visual guidance of grasping (e.g., Le, Vesia, Yan, Niemeier, & Crawford, 2014; Rice, Tunik, & Grafton, 2006; Tunik, Frey, & Grafton, 2005) and posterior IPS in the visual guidance of reaching (e.g., Striemer, Chouinard, & Goodale, 2011; Vesia, Prime, Yan, Sergio, & Crawford, 2010; Vesia, Yan, Henriques, Sergio, & Crawford, 2008; Vesia, Monteon, Sergio, & Crawford, 2006). However, as far as we know, no single TMS study has yet demonstrated a complete double dissociation.

Reprint requests should be sent to Philippe A. Chouinard, School of Psychology and Public Health, Applied Science 2 Building, Room 3.14, La Trobe University, Bendigo Campus, Bendigo, Victoria 3550, Australia, or via e-mail: p.chouinard@latrobe.edu.au.

REFERENCES

Bar
,
M.
,
Kassam
,
K. S.
,
Ghuman
,
A. S.
,
Boshyan
,
J.
,
Schmid
,
A. M.
,
Dale
,
A. M.
, et al
(
2006
).
Top--down facilitation of visual recognition
.
Proceedings of the National Academy of Sciences, U.S.A.
,
103
,
449
454
.
Biederman
,
I.
(
1987
).
Recognition-by-components: A theory of human image understanding
.
Psychological Review
,
94
,
115
147
.
Bridge
,
H.
,
Thomas
,
O. M.
,
Minini
,
L.
,
Cavina-Pratesi
,
C.
,
Milner
,
A. D.
, &
Parker
,
A. J.
(
2013
).
Structural and functional changes across the visual cortex of a patient with visual form agnosia
.
Journal of Neuroscience
,
33
,
12779
12791
.
Bub
,
D. N.
, &
Lewine
,
J.
(
1988
).
Different modes of word recognition in the left and right visual fields
.
Brain and Language
,
33
,
161
188
.
Chiarello
,
C.
,
Nuding
,
S.
, &
Pollock
,
A.
(
1988
).
Lexical decision and naming asymmetries: Influence of response selection and response bias
.
Brain and Language
,
34
,
302
314
.
Chouinard
,
P. A.
, &
Goodale
,
M. A.
(
2010
).
Category-specific neural processing for naming pictures of animals and naming pictures of tools: An ALE meta-analysis
.
Neuropsychologia
,
48
,
409
418
.
Chouinard
,
P. A.
, &
Paus
,
T.
(
2010
).
What have we learned from “perturbing” the human cortical motor system with transcranial magnetic stimulation?
Frontiers in Human Neuroscience
,
4
,
173
.
Chouinard
,
P. A.
,
Whitwell
,
R. L.
, &
Goodale
,
M. A.
(
2009
).
The lateral-occipital and the inferior-frontal cortex play different roles during the naming of visually presented objects
.
Human Brain Mapping
,
30
,
3851
3864
.
Cohen
,
J.
(
1988
).
Statistical power analysis for the behavioral sciences
(2nd ed.).
Hillsdale, NJ
:
Lawrence Erlbaum Associates
.
Cohen
,
L. G.
,
Roth
,
B. J.
,
Nilsson
,
J.
,
Dang
,
N.
,
Panizza
,
M.
,
Bandinelli
,
S.
, et al
(
1990
).
Effects of coil design on delivery of focal magnetic stimulation. Technical considerations
.
Electroencephalography and Clinical Neurophysiology
,
75
,
350
357
.
Corbetta
,
M.
, &
Shulman
,
G. L.
(
2002
).
Control of goal-directed and stimulus-driven attention in the brain
.
Nature Reviews Neuroscience
,
3
,
201
215
.
Ditunno
,
P. L.
, &
Mann
,
V. A.
(
1990
).
Right hemisphere specialization for mental rotation in normals and brain damaged subjects
.
Cortex
,
26
,
177
188
.
Ellison
,
A.
, &
Cowey
,
A.
(
2006
).
TMS can reveal contrasting functions of the dorsal and ventral visual processing streams
.
Experimental Brain Research
,
175
,
618
625
.
Farah
,
M. J.
(
2004
).
Visual agnosia
(2nd ed.).
Cambridge, MA
:
MIT Press
.
Finkbeiner
,
M.
,
Almeida
,
J.
, &
Caramazza
,
A.
(
2006
).
Letter identification processes in reading: Distractor interference reveals an automatically engaged, domain-specific mechanism
.
Cognitive Neuropsychology
,
23
,
1083
1103
.
Foxe
,
J. J.
, &
Simpson
,
G. V.
(
2002
).
Flow of activation from V1 to frontal cortex in humans. A framework for defining “early” visual processing
.
Experimental Brain Research
,
142
,
139
150
.
Goodale
,
M. A.
, &
Milner
,
A. D.
(
1992
).
Separate visual pathways for perception and action
.
Trends in Neurosciences
,
15
,
20
25
.
Goodale
,
M. A.
,
Milner
,
A. D.
,
Jakobson
,
L. S.
, &
Carey
,
D. P.
(
1991
).
A neurological dissociation between perceiving objects and grasping them
.
Nature
,
349
,
154
156
.
Grill-Spector
,
K.
,
Kushnir
,
T.
,
Edelman
,
S.
,
Avidan
,
G.
,
Itzchak
,
Y.
, &
Malach
,
R.
(
1999
).
Differential processing of objects under various viewing conditions in the human lateral occipital complex
.
Neuron
,
24
,
187
203
.
Harris
,
I. M.
,
Benito
,
C. T.
,
Ruzzoli
,
M.
, &
Miniussi
,
C.
(
2008
).
Effects of right parietal transcranial magnetic stimulation on object identification and orientation judgments
.
Journal of Cognitive Neuroscience
,
20
,
916
926
.
Harris
,
I. M.
,
Harris
,
J. A.
, &
Caine
,
D.
(
2001
).
Object orientation agnosia: A failure to find the axis?
Journal of Cognitive Neuroscience
,
13
,
800
812
.
Harris
,
I. M.
, &
Miniussi
,
C.
(
2003
).
Parietal lobe contribution to mental rotation demonstrated with rTMS
.
Journal of Cognitive Neuroscience
,
15
,
315
323
.
Hubel
,
D. H.
, &
Wiesel
,
T. N.
(
1968
).
Receptive fields and functional architecture of monkey striate cortex
.
Journal of Physiology
,
195
,
215
243
.
Hunter
,
Z. R.
, &
Brysbaert
,
M.
(
2008
).
Visual half-field experiments are a good measure of cerebral language dominance if used properly: Evidence from fMRI
.
Neuropsychologia
,
46
,
316
325
.
Inui
,
K.
, &
Kakigi
,
R.
(
2006
).
Temporal analysis of the flow from V1 to the extrastriate cortex in humans
.
Journal of Neurophysiology
,
96
,
775
784
.
James
,
T. W.
,
Culham
,
J.
,
Humphrey
,
G. K.
,
Milner
,
A. D.
, &
Goodale
,
M. A.
(
2003
).
Ventral occipital lesions impair object recognition but not object-directed grasping: An fMRI study
.
Brain
,
126
,
2463
2475
.
James
,
T. W.
,
Humphrey
,
G. K.
,
Gati
,
J. S.
,
Menon
,
R. S.
, &
Goodale
,
M. A.
(
2002
).
Differential effects of viewpoint on object-driven activation in dorsal and ventral streams
.
Neuron
,
35
,
793
801
.
Jonides
,
J.
,
Smith
,
E. E.
,
Koeppe
,
R. A.
,
Awh
,
E.
,
Minoshima
,
S.
, &
Mintun
,
M. A.
(
1993
).
Spatial working memory in humans as revealed by PET
.
Nature
,
363
,
623
625
.
Jordan
,
K.
,
Heinze
,
H. J.
,
Lutz
,
K.
,
Kanowski
,
M.
, &
Jancke
,
L.
(
2001
).
Cortical activations during the mental rotation of different visual objects
.
Neuroimage
,
13
,
143
152
.
Karnath
,
H. O.
(
1997
).
Spatial orientation and the representation of space with parietal lobe lesions
.
Philosophical Transactions of the Royal Society of London, Series B, Biological Sciences
,
352
,
1411
1419
.
Knecht
,
S.
,
Drager
,
B.
,
Deppe
,
M.
,
Bobe
,
L.
,
Lohmann
,
H.
,
Floel
,
A.
, et al
(
2000
).
Handedness and hemispheric language dominance in healthy humans
.
Brain
,
123
,
2512
2518
.
Kourtzi
,
Z.
,
Erb
,
M.
,
Grodd
,
W.
, &
Bulthoff
,
H. H.
(
2003
).
Representation of the perceived 3-D object shape in the human lateral occipital complex
.
Cerebral Cortex
,
13
,
911
920
.
Larsson
,
J.
, &
Heeger
,
D. J.
(
2006
).
Two retinotopic visual areas in human lateral occipital cortex
.
Journal of Neuroscience
,
26
,
13128
13142
.
Larsson
,
J.
,
Landy
,
M. S.
, &
Heeger
,
D. J.
(
2006
).
Orientation-selective adaptation to first- and second-order patterns in human visual cortex
.
Journal of Neurophysiology
,
95
,
862
881
.
Laycock
,
R.
,
Crewther
,
S. G.
, &
Crewther
,
D. P.
(
2007
).
A role for the “magnocellular advantage” in visual impairments in neurodevelopmental and psychiatric disorders
.
Neuroscience & Biobehavioral Reviews
,
31
,
363
376
.
Le
,
A.
,
Vesia
,
M.
,
Yan
,
X.
,
Niemeier
,
M.
, &
Crawford
,
J. D.
(
2014
).
The right anterior intraparietal sulcus is critical for bimanual grasping: A TMS study
.
Cerebral Cortex
,
24
,
2591
2603
.
Maccabee
,
P. J.
,
Eberle
,
L.
,
Amassian
,
V. E.
,
Cracco
,
R. Q.
,
Rudell
,
A.
, &
Jayachandra
,
M.
(
1990
).
Spatial distribution of the electric field induced in volume by round and figure ‘8’ magnetic coils: Relevance to activation of sensory nerve fibers
.
Electroencephalography and Clinical Neurophysiology
,
76
,
131
141
.
Milner
,
A. D.
, &
Goodale
,
M. A.
(
2008
).
Two visual systems re-viewed
.
Neuropsychologia
,
46
,
774
785
.
Milner
,
A. D.
,
Perrett
,
D. I.
,
Johnston
,
R. S.
,
Benson
,
P. J.
,
Jordan
,
T. R.
,
Heeley
,
D. W.
, et al
(
1991
).
Perception and action in “visual form agnosia”
.
Brain
,
114
,
405
428
.
Mountcastle
,
V. B.
(
1957
).
Modality and topographic properties of single neurons of cat's somatic sensory cortex
.
Journal of Neurophysiology
,
20
,
408
434
.
Pascual-Leone
,
A.
,
Walsh
,
V.
, &
Rothwell
,
J.
(
2000
).
Transcranial magnetic stimulation in cognitive neuroscience—Virtual lesion, chronometry, and functional connectivity
.
Current Opinion in Neurobiology
,
10
,
232
237
.
Paus
,
T.
(
2005
).
Inferring causality in brain images: A perturbation approach
.
Philosophical Transactions of the Royal Society of London, Series B, Biological Sciences
,
360
,
1109
1114
.
Pitcher
,
D.
,
Charles
,
L.
,
Devlin
,
J. T.
,
Walsh
,
V.
, &
Duchaine
,
B.
(
2009
).
Triple dissociation of faces, bodies, and objects in extrastriate cortex
.
Current Biology
,
19
,
319
324
.
Rice
,
N. J.
,
Tunik
,
E.
, &
Grafton
,
S. T.
(
2006
).
The anterior intraparietal sulcus mediates grasp execution, independent of requirement to update: New insights from transcranial magnetic stimulation
.
Journal of Neuroscience
,
26
,
8176
8182
.
Russell
,
C.
,
Deidda
,
C.
,
Malhotra
,
P.
,
Crinion
,
J. T.
,
Merola
,
S.
, &
Husain
,
M.
(
2010
).
A deficit of spatial remapping in constructional apraxia after right-hemisphere stroke
.
Brain
,
133
,
1239
1251
.
Sacks
,
O.
(
1985
).
The man who mistook his wife for a hat and other clinical tales
.
New York
:
Summit Books
.
Sakata
,
H.
,
Taira
,
M.
,
Kusunoki
,
M.
,
Murata
,
A.
,
Tanaka
,
Y.
, &
Tsutsui
,
K.
(
1998
).
Neural coding of 3D features of objects for hand action in the parietal cortex of the monkey
.
Philosophical Transactions of the Royal Society of London, Series B, Biological Sciences
,
353
,
1363
1373
.
Sandrini
,
M.
,
Umilta
,
C.
, &
Rusconi
,
E.
(
2011
).
The use of transcranial magnetic stimulation in cognitive neuroscience: A new synthesis of methodological issues
.
Neuroscience & Biobehavioral Reviews
,
35
,
516
536
.
Saxe
,
R.
,
Brett
,
M.
, &
Kanwisher
,
N.
(
2006
).
Divide and conquer: A defense of functional localizers
.
Neuroimage
,
30
,
1088
1096
.
Sergent
,
J.
(
1983
).
Role of the input in visual hemispheric asymmetries
.
Psychological Bulletin
,
93
,
481
512
.
Shikata
,
E.
,
Hamzei
,
F.
,
Glauche
,
V.
,
Knab
,
R.
,
Dettmers
,
C.
,
Weiller
,
C.
, et al
(
2001
).
Surface orientation discrimination activates caudal and anterior intraparietal sulcus in humans: An event-related fMRI study
.
Journal of Neurophysiology
,
85
,
1309
1314
.
Shikata
,
E.
,
Hamzei
,
F.
,
Glauche
,
V.
,
Koch
,
M.
,
Weiller
,
C.
,
Binkofski
,
F.
, et al
(
2003
).
Functional properties and interaction of the anterior and posterior intraparietal areas in humans
.
European Journal of Neuroscience
,
17
,
1105
1110
.
Silson
,
E. H.
,
McKeefry
,
D. J.
,
Rodgers
,
J.
,
Gouws
,
A. D.
,
Hymers
,
M.
, &
Morland
,
A. B.
(
2013
).
Specialized and independent processing of orientation and shape in visual field maps LO1 and LO2
.
Nature Neuroscience
,
16
,
267
269
.
Striemer
,
C. L.
,
Chouinard
,
P. A.
, &
Goodale
,
M. A.
(
2011
).
Programs for action in superior parietal cortex: A triple-pulse TMS investigation
.
Neuropsychologia
,
49
,
2391
2399
.
Taira
,
M.
,
Tsutsui
,
K. I.
,
Jiang
,
M.
,
Yara
,
K.
, &
Sakata
,
H.
(
2000
).
Parietal neurons represent surface orientation from the gradient of binocular disparity
.
Journal of Neurophysiology
,
83
,
3140
3146
.
Tsutsui
,
K.
,
Sakata
,
H.
,
Naganuma
,
T.
, &
Taira
,
M.
(
2002
).
Neural correlates for perception of 3D surface orientation from texture gradient
.
Science
,
298
,
409
412
.
Tukey
,
J.
(
1949
).
Comparing individual means in the analysis of variance
.
Biometrics
,
5
,
99
114
.
Tunik
,
E.
,
Frey
,
S. H.
, &
Grafton
,
S. T.
(
2005
).
Virtual lesions of the anterior intraparietal area disrupt goal-dependent on-line adjustments of grasp
.
Nature Neuroscience
,
8
,
505
511
.
Turnbull
,
O. H.
,
Beschin
,
N.
, &
Della Sala
,
S.
(
1997
).
Agnosia for object orientation: Implications for theories of object recognition
.
Neuropsychologia
,
35
,
153
163
.
Turnbull
,
O. H.
,
Laws
,
K. R.
, &
McCarthy
,
R. A.
(
1995
).
Object recognition without knowledge of object orientation
.
Cortex
,
31
,
387
395
.
Urgesi
,
C.
,
Candidi
,
M.
,
Ionta
,
S.
, &
Aglioti
,
S. M.
(
2007
).
Representation of body identity and body actions in extrastriate body area and ventral premotor cortex
.
Nature Neuroscience
,
10
,
30
31
.
Valyear
,
K. F.
,
Culham
,
J. C.
,
Sharif
,
N.
,
Westwood
,
D.
, &
Goodale
,
M. A.
(
2006
).
A double dissociation between sensitivity to changes in object identity and object orientation in the ventral and dorsal visual streams: A human fMRI study
.
Neuropsychologia
,
44
,
218
228
.
Vesia
,
M.
,
Monteon
,
J. A.
,
Sergio
,
L. E.
, &
Crawford
,
J. D.
(
2006
).
Hemispheric asymmetry in memory-guided pointing during single-pulse transcranial magnetic stimulation of human parietal cortex
.
Journal of Neurophysiology
,
96
,
3016
3027
.
Vesia
,
M.
,
Prime
,
S. L.
,
Yan
,
X.
,
Sergio
,
L. E.
, &
Crawford
,
J. D.
(
2010
).
Specificity of human parietal saccade and reach regions during transcranial magnetic stimulation
.
Journal of Neuroscience
,
30
,
13053
13065
.
Vesia
,
M.
,
Yan
,
X.
,
Henriques
,
D. Y.
,
Sergio
,
L. E.
, &
Crawford
,
J. D.
(
2008
).
Transcranial magnetic stimulation over human dorsal-lateral posterior parietal cortex disrupts integration of hand position signals into the reach plan
.
Journal of Neurophysiology
,
100
,
2005
2014
.
Vingerhoets
,
G.
(
2008
).
Knowing about tools: Neural correlates of tool familiarity and experience
.
Neuroimage
,
40
,
1380
1391
.
Vuilleumier
,
P.
,
Henson
,
R. N.
,
Driver
,
J.
, &
Dolan
,
R. J.
(
2002
).
Multiple levels of visual object constancy revealed by event-related fMRI of repetition priming
.
Nature Neuroscience
,
5
,
491
499
.
Wassermann
,
E. M.
(
1998
).
Risk and safety of repetitive transcranial magnetic stimulation: Report and suggested guidelines from the International Workshop on the Safety of Repetitive Transcranial Magnetic Stimulation, June 5–7, 1996
.
Electroencephalography and Clinical Neurophysiology
,
108
,
1
16
.
Wassermann
,
E. M.
,
Wang
,
B.
,
Zeffiro
,
T. A.
,
Sadato
,
N.
,
Pascual-Leone
,
A.
,
Toro
,
C.
, et al
(
1996
).
Locating the motor cortex on the MRI with transcranial magnetic stimulation and PET
.
Neuroimage
,
3
,
1
9
.
Whitwell
,
R. L.
,
Milner
,
A. D.
, &
Goodale
,
M. A.
(
2014
).
The two visual systems hypothesis: New challenges and insights from visual form agnosic patient DF
.
Frontiers in Neurology
,
5
,
255
.
Wood
,
D. K.
,
Chouinard
,
P. A.
,
Major
,
A. J.
, &
Goodale
,
M. A.
(
in press
).
Sensitivity to biomechanical limitations during postural decision-making depends on the integrity of posterior superior parietal cortex
.
Cortex
.
doi: 10.1016/j.cortex.2016.07.005
.
Worsley
,
K. J.
,
Evans
,
A. C.
,
Marrett
,
S.
, &
Neelin
,
P.
(
1992
).
Determining the number of statistically significant areas of activation in subtracted activation studies from PET
.
Journal of Cerebral Blood Flow and Metabolism
,
12
,
900
918
.