- Email: [email protected]
Differential Effects of Transcranial Magnetic Stimulation of Left and Right Posterior Parietal Cortex on Mental Rotation Tasks
SPECIAL SECTION DIFFERENTIAL EFFECTS OF TRANSCRANIAL MAGNETIC STIMULATION OF LEFT AND RIGHT POSTERIOR PARIETAL CORTEX ON MENTAL ROTATION TASKS Eva A. Feredoes1,2 and Perminder S. Sachdev1,2 (1School of Psychiatry, Faculty of Medicine, University of New South Wales, Sydney, Australia; 2Neuropsychiatric Institute, Prince of Wales Hospital, Sydney, Australia)
ABSTRACT A recently published study used the interference strategy of transcranial magnetic stimulation (TMS) to demonstrate the role of the right posterior parietal cortex (PPC) in the mental rotation of alphanumeric stimuli. We used similar stimulation parameters over the same left and right PPC regions, and examined the ability to rotate more complex 3D Shepard and Metzler (1971) images. There was reduced accuracy of performance with both right and left PPC stimulation for different angles of rotation of the visual stimuli. Right PPC stimulation led to reduced accuracy to rotate stimuli by 120°, whereas left PPC stimulation affected 180° rotation. We hypothesise that the two hemispheres make different contributions to the processing underlying visuospatial mental imagery: the right PPC is important for spatial rotations through smaller angles; the left hemisphere has a unique role when the stimuli to be compared are rotated through 180°, a task that engages verbal strategies due to the well-documented special nature of enantiomorphs. Key words: TMS, Shepard-Metzler, posterior parietal, enantiomorphs
INTRODUCTION Mental rotation is a well-studied paradigm of imagery. The landmark investigation by Shepard and Metzler (1971) showed that response times increased linearly as a function of angular disparity, a robust effect that has been replicated on many occasions. Based on their observations, Shepard and Metzler (1971) proposed that mental rotation involves subjects imagining a rotated shape passing through intermediate orientations until it matches the target. The larger the rotation angle from upright, the greater the required rotation and the longer the response times (Shepard and Metzler, 1971). Human lesion studies (Mehta and Newcombe, 1991; Ratcliff, 1979), animal investigations (Ockleford et al., 1977) and neuroimaging techniques (Alivisatos and Petrides, 1997; Cohen et al., 1996; Harris and Miniussi, 2003; Harris et al., 2000; Pegna et al., 1997) have all contributed to the evidence that the posterior parietal cortex (PPC) plays an important role in carrying out processes required for mental rotation. The differential contribution of the left relative to the right cerebral hemisphere is, however, a topic of debate, and the extensive discussion by Harris et al. (2000) suggests that methodological differences between neuroimaging techniques and experimental procedural differences in the tasks used may account for differences in activation patterns between published studies. In this paper, we address the issue of lateralisation of processing Cortex, (2006) 42, 750-754
with the interference methodology of transcranial magnetic stimulation (TMS). TMS can disrupt the processing of a targeted brain area for a limited period. TMS induces an electric current in a small region of the brain by delivering a rapidly varying high magnetic field produced by a stimulating coil held over the targeted brain region. The resulting disruption of activity in the targeted brain region leads to an alteration in, or suspension of, behaviour consequent upon that brain activity. This “virtual lesion” allows the attribution of a process to an anatomical location (Walsh and Pascual-Leone, 2003). The aim of the present investigation was to confirm the involvement of the right PPC in mental rotation (Harris and Miniussi, 2003; Harris et al., 2000) and also to investigate whether and to what extent the left PPC is recruited for the rotation of more complex Shepard and Metzler (1971) stimuli, as predicted by Harris and Miniussi (2003) and Carpenter et al. (1999). MATERIALS AND METHODS Subjects Twenty healthy right-handed subjects were recruited from the general population (mean age = 24.7 years, 10 males and 10 females). They provided written informed consent and received a small payment for their participation and the study was approved by the institutional ethics committee of the University of New South Wales.
Differential effects of rTMS on mental rotation
Fig. 1 – One stimulus presentation trial. A pair of 3D stimuli was presented in the centre of a computer monitor for 2500 msec during which subjects were to identify the rotated stimulus on the right as the same or a mirror-image version of the stimulus on the left. 400 msec after stimulus onset, 4 magnetic pulses were given for a duration of 200 msec. After the stimulus disappeared, a 3500 msec inter-trial interval followed. There were 38 trials in a block.
TMS A MagPro X100 stimulator (Medtronic, Skovlunde, Denmark) was used to deliver pulses at 60% of maximum machine intensity (di/dt = 92 µA/msec). A fluid-cooled figure-of-eight coil with 50 mm loop diameter was used. Four pulses at 20 Hz were time-locked to be delivered 400 msec after onset of presentation of the visual stimuli (at 400, 450, 500 and 550 msec). A timeline of one experimental trial is shown in Figure 1. The site of stimulation was located according to the method used by Harris and Miniussi (2003) where the PPC was approximated to lie under the point halfway between CP4 and P4, and CP3 and P3 on the international 10-20 system of electroencephalography (EEG) montages. Subjects wore a custom-designed neoprene cap with these points marked out, making repeated coil localisation more reliable. Each subject received 304 pulses in total. The coil was positioned throughout each trial with the handle pointing in an anterior direction. Subjects reported no significant discomfort from the stimulation. Sham stimulation was given to the left and right sites with the coil held perpendicular to the scalp, with the edge of one wing in contact with the scalp surface. Task Eight white 3D figures taken from Shepard and Metzler (1971) were presented on a black
751
background. Pairs of stimuli were presented on a 15" monitor (60 Hz refresh rate), with the stimulus on the left being the target to which the right-sided stimulus was matched. The stimulus on the right could be rotated 0°, 60°, 120° or 180° and subjects were instructed to rotate the stimulus to the “upright” position of the left stimulus and then judge if it is the same or a mirror of the target. The stimuli were rotated in depth, as per Shepard and Metzler (1971) where a non-rigid transformation was used to correspond to a rigid rotation of a 3D object. Subjects were seated 60 cm from the monitor (visual angle ~ 5°) and pressed a key on a computer keyboard with their right index finger for a normal stimulus and with their right thumb for a mirror presentation. Stimuli were presented for 2500 msec followed by a 3500 msec interval with a blank screen. This inter-trial interval was designed to comply with the TMS safety recommendations of Wasserman (1998). Subjects were instructed to respond as quickly and accurately as possible whilst the stimuli remained on-screen and any response made after this was scored as incorrect. A total of 38 stimuli were presented in a block, lasting for 3 min and 48 sec. In order to minimise testing time and to allow the subjects to become familiar with the task, a practice phase was performed one day prior to testing. Under testing, the task was performed four times in total, during left and right active and sham stimulation and stimulation order was counterbalanced across subjects. Stimulation was given over one site/task block and accuracies and response times were recorded. RESULTS Our data showed the typical trend first observed by Shepard and Metzler (1971) that reaction time increases linearly with angular disparity. The exception to this finding was for 180° rotation, for which reaction times did not increase from those rotated by 120°, under either real or sham stimulation. No significant main effects of stimulation were evident on the response times of subjects, using a repeated measures analysis of variance (ANOVA) [F (1, 18) = 1.408, p = .251], and no significant interaction occurred between type and side of stimulation and angle of rotation [F (3, 54) = .466, p = .220]. The results of the reaction times are presented in Figure 2. To examine the effect of repetititve transcranial magnetic stimulation (rTMS) on accuracy of performance, a repeated-measures ANOVA with stimulation type (real, sham) × side (left, right) × angle (0°, 60°, 120°, 180°) as within-subject factors found a significant main effect of stimulation [F (1, 18) = 5.489, p = .031]. There was, as expected, a significant difference in
752
Eva A. Feredoes and Perminder S. Sachdev
Fig. 2 – Mean reaction times under each condition, which increased with angular disparity, except for inverted (180°) stimuli. The TMS conditions did not differ significantly from sham conditions. Error bars indicate ± 1 SEM.
accuracies between the angles of rotation [F (3, 54) = 56.968, p = .0001]. A significant interaction was evident between the type and side of stimulation and the angle of rotation [F (3, 54) = 3.110, p = .034]. Pairwise comparisons showed that right PPC TMS reduced accuracy for 60° rotation nonsignificantly (p = .087; Bonferroni corrected), and for 120° significantly (p = .003; Bonferroni corrected) but not for 180° rotation (Figure 3). Left-sided stimulation on the other hand, significantly reduced accuracy for 180° rotation (p = .012; Bonferroni corrected), but not for the other angles examined (Figure 4). DISCUSSION
Fig. 3 – Mean accuracy under right PPC TMS compared to sham. Stimulation significantly decreased accuracy for 120°rotated angles. Error bars indicate ± 1 SEM.
Fig. 4 – Mean accuracy under left PPC TMS compared to sham. Stimulation significantly decreased accuracy for 180°rotated angles. Error bars indicate ± 1 SEM.
The aims of this investigation were to confirm the contribution of the right PPC and explore possible involvement of the left hemisphere in the mental rotation of complex 3D stimuli. Harris and Miniussi (2003) have previously demonstrated that mental rotation was disrupted when TMS was delivered over the right PPC, 400-600 msec poststimulus onset. They hypothesised that the more complex Shepard and Metzler (1971) shapes may recruit the left PPC, and we set out to test this using TMS. Our results replicated the findings of Harris and Miniussi (2003) that right-sided TMS disrupted task performance, which was statistically significant for stimuli rotated by 120°. This is further evidence for the central role this region plays in the processing of spatial transformations required for mental rotation. Stimuli rotated by 120° demand a comparatively larger amount of mental processing, as reflected by the increased errors and response times under control conditions, and are therefore particularly susceptible to interference. TMS also reduced the accuracy of 60° rotation but this did not reach significance. TMSinduced performance deficits are more likely to occur when visual stimulus presentation is brief, or close to threshold of performance (Walsh and Pascual-Leone, 2003). In this study the duration of
Differential effects of rTMS on mental rotation
TMS was very short relative to the long stimulus viewing times. This is also likely to have masked any effects of TMS on reaction time because responses were not under time pressure (see Walsh and Pascual-Leone, 2003, for a full methodological discussion). It could also be argued that the effects produced by TMS can be explained by task difficulty, disrupting only the stimuli with the longest reaction times. This, however, is unlikely as left and right PPC TMS had different effects on stimuli that had similar reaction times. Interestingly, left PPC stimulation produced a significant decrease in accuracy for inverted stimuli (180° rotation). The result indicates a role for this region in mental rotation that is different from the role played by the right PPC. A closer examination of the special nature of 180° rotations may offer an explanation. This observation that not all angles of rotation are processed equally has been shown with respect to inverted stimuli (Turnbull et al., 2004; Walsh and Butler, 1996) where the increase in response times do not follow the expected linear pattern from 120° to 180°. Our behavioural results replicate this common finding. As vertically inverted stimuli are often simply an upside-down version of the target, Cooper and Shepard (1973) proposed that instead of piecemeal rotation used for smaller angles of rotation, subjects flip the whole object about the horizontal axis and then about the vertical (if mirrored; see also Walsh and Butler, 1996 and Carpenter et al., 1999) and response times are increased when a lateral plus vertical rotation must be made, compared to only vertical rotations. Different processing mechanisms between rotated and inverted shapes have also been reported in the monkey lesion literature and human neuropsychological literature. Parietal lesions in monkeys produced deficits in recognising shapes that had been left-right inverted but not for updown inversions (Ockleford et al., 1977). A left temporo-parietal lesion in a human subject showed the same pattern of results (Riddoch and Humphreys, 1988). It has been suggested that in the natural environment, left-right orientation is not important for discriminating between two objects, as this usually reflects the vertical symmetry natural to faces, front views of bodies and many living organisms (Gross and Bornstein, 1978). These enantiomorphic forms are therefore regarded by the brain as perceptually equivalent (Gross and Bornstein, 1978) and differentiating between two such shapes requires a greater degree of spatial processing (Walsh and Butler, 1996). Indeed, the unusual nature of enantiomorphs has been suggested as an explanation for the many cells in the inferotemporal cortex which respond to objects particularly strongly for their mirror inversions (Logothetis et al., 1995). According to Walsh and Butler (1996),
753
recognition of vertically inverted stimuli is carried out by areas more involved in object rather than spatial processing. This is supported by our results which show that under right-sided stimulation, the right side being pre-eminent in visuospatial processing; there is no effect of TMS on inverted stimuli specifically. Only left-sided stimulation was able to significantly disrupt these stimuli. The involvement of the left hemisphere in visual imagery is well-established, specifically for the verbal categorical encoding of an object (Kosslyn et al., 1995; Michimata, 1997; Parrot et al., 1999). Spatial relations can be represented by categories that can be easily verbalised, such as “above/below, left to/right to”, or alternatively, by a coordinate strategy that encodes the metric information of an object (Kosslyn et al., 1995). Neuroimaging studies have found preferential left parietal activation for categorical tasks, whilst the right is more active when precise spatial information is utilised (Baciu et al., 1999; Trojano et al., 2002). It is possible that inverted stimuli were processed using verbally-based categorical information, mediated by the left PPC and as a consequence, were the only such stimuli to be disrupted by left-sided stimulation. A role of the left parietal cortex in mental rotation has been difficult to establish. Alivisatos and Petrides (1997) found left parietal activity specific to mental rotation of alphanumeric characters, as did Vingerhoets et al. (2001) for paired 2D stimuli. Other studies have reported bilateral parietal involvement (Carpenter et al., 1999; Kosslyn et al., 1998; Richter et al., 1997; Tagaris et al., 1997) or only right hemispheric processing (Deutsch et al., 1988; Harris et al., 2000). These differences could be due to procedural differences between investigations, such as the different range of stimuli orientations, instructions given, task practice, or the type of neuroimaging approach used (Harris et al., 2000). Corballis (1997) suggested left-hemisphere involvement in mental rotation tasks utilising more complex stimuli due to subjects’ use of verbal or sequential strategies. Neuro- and experimental psychology evidence is available to support this (Corballis, 1997; McGuinness and Bartell, 1982; Mehta and Newcombe, 1991) and subjective reports from normal subjects are that verbal or symbolic strategies are used when rotating complex 3D (but not 2D) figures (McGuinness and Bartell, 1982). Monkey lesion evidence is also available showing that stimuli of different dimensions are not processed in the same way (Holmes and Gross, 1984). CONCLUSION Our investigation supports the important role of right PPC in mental rotation. It also shows that the left PPC plays an important and hitherto under-
754
Eva A. Feredoes and Perminder S. Sachdev
investigated role in mental rotation, especially when it involves inversion of a 3D figure. Acknowledgements. We are incredibly grateful to Vincent Walsh for his expert comments on the manuscript and to Garth Hargreaves for developing the computerised stimuli. REFERENCES ALIVISATOS B and PETRIDES M. Functional activation of the human brain during mental rotation. Neuropsychologia, 35: 111-118, 1997. BACIU M, KOENIG O, VERNIER MP, BEDOIN N, RUBIN C and SEGEBARTH C. Categorical and coordinate spatial relations: fMRI evidence for hemispheric specialization. Neuroreport, 10: 1373-1378, 1999. CARPENTER PA, JUST MA, KELLER TA and EDDY W. Graded functional activation in the visuospatial system with the amount of task demand. Journal of Cognitive Neuroscience, 11: 9-24, 1999. COHEN MS, KOSSLYN SM, BREITER HC, DIGIROLAMO GJ, THOMPSON WL, ANDERSON AK, BOOKHEIMER SY, ROSEN BR and BENNETT J. Changes in cortical activity during mental rotation: A mapping study using functional MRI. Brain, 119: 89-100, 1996. COOPER LA and SHEPARD RN. Chronometric studies of the rotation of mental images. In Chase WG (Ed), Visual Information Processing. New York: Academic Press, 1973. CORBALLIS MC. Mental rotation and the right hemisphere. Brain and Language, 57: 100-121, 1997. DEUTSCH G, BOURBON WT, PAPANICOLAOU AC and EISENBERG HM. Visuospatial tasks compared via activation of regional cerebral blood flow. Neuropsychologia, 26: 445-452, 1988. GROSS CG and BORNSTEIN MH. Left and right in science and art. Leonardo, 11: 29-38, 1978. HARRIS IM, EGAN GF, SONKKILA C, TOCHON-DANGUY HJ, PAXINOS G and WATSON JDG. Selective right parietal lobe activation during mental rotation: A parametric PET study. Brain, 123: 65-73, 2000. HARRIS IM and MINIUSSI C. Parietal lobe contribution to mental rotation demonstrated with rTMS. Journal of Cognitive Neuroscience, 15: 315-323, 2003. HOLMES EJ and GROSS CG. Stimulus equivalence after inferior temporal lesions in monkeys. Behavioral Neuroscience, 98: 898-901, 1984. KOSSLYN SM, DIGIROLAMO GJ, THOMPSON WL and ALPERT NM. Mental rotation of objects versus hands: Neural mechanisms revealed by positron emission tomography. Psychophysiology, 35: 151-161, 1998. KOSSLYN SM, MALJKOVIC V, HAMILTON SE, HORWITZ G and THOMPSON WL. Two types of image generation: Evidence for left and right hemispheric processes. Neuropsychologia, 33: 1485-1510, 1995.
LOGOTHETIS NK, PAULS J and POGGIO T. Shape representation in the inferior temporal cortex of monkeys. Current Biology, 5: 552-563, 1995. MCGUINNESS D and BARTELL DE. Hard or simple-minded? Neuropsychologia, 20: 629-639, 1982. MEHTA Z and NEWCOMBE F. A role for the left hemisphere in spatial processing. Cortex, 23: 163-167, 1991. MICHIMATA C. Hemispheric processing of categorical and coordinate spatial relations in vision and visual imagery. Brain and Cognition, 33: 370-387, 1997. OCKLEFORD EM, MILNER AD, DEWAR W and SNEDDON IA. Form perception in stumptail macaques following posterior parietal and lateral frontal lesions. Cortex, 13: 361-372, 1977. PARROT M, DOYON B, DEMONET JF and CARDEBAT D. Hemispheric preponderance in categorical and coordinate visual processes. Neuropsychologia, 37: 1215-1225, 1999. PEGNA AJ, KHATEB A, SPINELLI L, SEECK M, LANDIS T and MICHEL CM. Unraveling the cerebral dynamics of mental imagery. Human Brain Mapping, 5: 410-421, 1997. RATCLIFF G. Spatial thought, mental rotation and the right cerebral hemisphere. Neuropsychologia, 17: 49-54, 1979. RICHTER W, UGURBIL K, GEORGOPOLOUS AP and KIM SG. Timeresolved fMRI of mental rotation. Neuroreport, 8: 3697-3702, 1997. RIDDOCH MJ and HUMPHREYS GW. Description of a left/right coding deficit in a case of constructional apraxia. Cognitive Neuropsychology, 5: 289-315, 1988. SHEPARD RN and METZLER J. Mental rotation of three-dimensional objects. Science, 19: 701-703, 1971. TAGARIS GA, KIM SG, STRUPP JP, ANDERSON P, UGURBIL K and GEORGOPOLOUS AP. Mental rotation studied by functional magnetic resonance imaging at high field (4T): Performance and cortical activation. Journal of Cognitive Neuroscience, 9: 419-432, 1997. TROJANO L, GROSSI D, LINDEN DE, FORMISANO E, GOEBEL R, CIRILLO S, ELEFANTE R and DI SALLE F. Coordinate and categorical judgments in spatial imagery: An fMRI study. Neuropsychologia, 40: 1666-1674, 2002. TURNBULL OH, DRIVER J and MCCARTHY RA. 2D but not 3D: Pictorial-depth deficits in a case of visual agnosia. Cortex, 40: 723-738, 2004. WALSH V and BUTLER SR. The effects of visual cortex lesions on the perception of rotated shapes. Behavioural Brain Research, 76: 127-142, 1996. WALSH V and PASCUAL-LEONE A. Transcranial magnetic stimulation: A neurochronometrics of mind. Cambridge, MA: MIT Press, 2003. W ASSERMAN E. Risk and safety of repetitive transcranial magnetic stimulation: report and suggested guidelines from the International Workshop in the Safety of Repetitive Transcranial Magnetic Stimulation. Electroencephalography and Clinical Neurophysiology, 108: 1-16, 1998.
Eva A. Feredoes, 1202 West Johnson Street, Madison WI 53706, USA. e-mail: [email protected]
ABSTRACT A recently published study used the interference strategy of transcranial magnetic stimulation (TMS) to demonstrate the role of the right posterior parietal cortex (PPC) in the mental rotation of alphanumeric stimuli. We used similar stimulation parameters over the same left and right PPC regions, and examined the ability to rotate more complex 3D Shepard and Metzler (1971) images. There was reduced accuracy of performance with both right and left PPC stimulation for different angles of rotation of the visual stimuli. Right PPC stimulation led to reduced accuracy to rotate stimuli by 120°, whereas left PPC stimulation affected 180° rotation. We hypothesise that the two hemispheres make different contributions to the processing underlying visuospatial mental imagery: the right PPC is important for spatial rotations through smaller angles; the left hemisphere has a unique role when the stimuli to be compared are rotated through 180°, a task that engages verbal strategies due to the well-documented special nature of enantiomorphs. Key words: TMS, Shepard-Metzler, posterior parietal, enantiomorphs
INTRODUCTION Mental rotation is a well-studied paradigm of imagery. The landmark investigation by Shepard and Metzler (1971) showed that response times increased linearly as a function of angular disparity, a robust effect that has been replicated on many occasions. Based on their observations, Shepard and Metzler (1971) proposed that mental rotation involves subjects imagining a rotated shape passing through intermediate orientations until it matches the target. The larger the rotation angle from upright, the greater the required rotation and the longer the response times (Shepard and Metzler, 1971). Human lesion studies (Mehta and Newcombe, 1991; Ratcliff, 1979), animal investigations (Ockleford et al., 1977) and neuroimaging techniques (Alivisatos and Petrides, 1997; Cohen et al., 1996; Harris and Miniussi, 2003; Harris et al., 2000; Pegna et al., 1997) have all contributed to the evidence that the posterior parietal cortex (PPC) plays an important role in carrying out processes required for mental rotation. The differential contribution of the left relative to the right cerebral hemisphere is, however, a topic of debate, and the extensive discussion by Harris et al. (2000) suggests that methodological differences between neuroimaging techniques and experimental procedural differences in the tasks used may account for differences in activation patterns between published studies. In this paper, we address the issue of lateralisation of processing Cortex, (2006) 42, 750-754
with the interference methodology of transcranial magnetic stimulation (TMS). TMS can disrupt the processing of a targeted brain area for a limited period. TMS induces an electric current in a small region of the brain by delivering a rapidly varying high magnetic field produced by a stimulating coil held over the targeted brain region. The resulting disruption of activity in the targeted brain region leads to an alteration in, or suspension of, behaviour consequent upon that brain activity. This “virtual lesion” allows the attribution of a process to an anatomical location (Walsh and Pascual-Leone, 2003). The aim of the present investigation was to confirm the involvement of the right PPC in mental rotation (Harris and Miniussi, 2003; Harris et al., 2000) and also to investigate whether and to what extent the left PPC is recruited for the rotation of more complex Shepard and Metzler (1971) stimuli, as predicted by Harris and Miniussi (2003) and Carpenter et al. (1999). MATERIALS AND METHODS Subjects Twenty healthy right-handed subjects were recruited from the general population (mean age = 24.7 years, 10 males and 10 females). They provided written informed consent and received a small payment for their participation and the study was approved by the institutional ethics committee of the University of New South Wales.
Differential effects of rTMS on mental rotation
Fig. 1 – One stimulus presentation trial. A pair of 3D stimuli was presented in the centre of a computer monitor for 2500 msec during which subjects were to identify the rotated stimulus on the right as the same or a mirror-image version of the stimulus on the left. 400 msec after stimulus onset, 4 magnetic pulses were given for a duration of 200 msec. After the stimulus disappeared, a 3500 msec inter-trial interval followed. There were 38 trials in a block.
TMS A MagPro X100 stimulator (Medtronic, Skovlunde, Denmark) was used to deliver pulses at 60% of maximum machine intensity (di/dt = 92 µA/msec). A fluid-cooled figure-of-eight coil with 50 mm loop diameter was used. Four pulses at 20 Hz were time-locked to be delivered 400 msec after onset of presentation of the visual stimuli (at 400, 450, 500 and 550 msec). A timeline of one experimental trial is shown in Figure 1. The site of stimulation was located according to the method used by Harris and Miniussi (2003) where the PPC was approximated to lie under the point halfway between CP4 and P4, and CP3 and P3 on the international 10-20 system of electroencephalography (EEG) montages. Subjects wore a custom-designed neoprene cap with these points marked out, making repeated coil localisation more reliable. Each subject received 304 pulses in total. The coil was positioned throughout each trial with the handle pointing in an anterior direction. Subjects reported no significant discomfort from the stimulation. Sham stimulation was given to the left and right sites with the coil held perpendicular to the scalp, with the edge of one wing in contact with the scalp surface. Task Eight white 3D figures taken from Shepard and Metzler (1971) were presented on a black
751
background. Pairs of stimuli were presented on a 15" monitor (60 Hz refresh rate), with the stimulus on the left being the target to which the right-sided stimulus was matched. The stimulus on the right could be rotated 0°, 60°, 120° or 180° and subjects were instructed to rotate the stimulus to the “upright” position of the left stimulus and then judge if it is the same or a mirror of the target. The stimuli were rotated in depth, as per Shepard and Metzler (1971) where a non-rigid transformation was used to correspond to a rigid rotation of a 3D object. Subjects were seated 60 cm from the monitor (visual angle ~ 5°) and pressed a key on a computer keyboard with their right index finger for a normal stimulus and with their right thumb for a mirror presentation. Stimuli were presented for 2500 msec followed by a 3500 msec interval with a blank screen. This inter-trial interval was designed to comply with the TMS safety recommendations of Wasserman (1998). Subjects were instructed to respond as quickly and accurately as possible whilst the stimuli remained on-screen and any response made after this was scored as incorrect. A total of 38 stimuli were presented in a block, lasting for 3 min and 48 sec. In order to minimise testing time and to allow the subjects to become familiar with the task, a practice phase was performed one day prior to testing. Under testing, the task was performed four times in total, during left and right active and sham stimulation and stimulation order was counterbalanced across subjects. Stimulation was given over one site/task block and accuracies and response times were recorded. RESULTS Our data showed the typical trend first observed by Shepard and Metzler (1971) that reaction time increases linearly with angular disparity. The exception to this finding was for 180° rotation, for which reaction times did not increase from those rotated by 120°, under either real or sham stimulation. No significant main effects of stimulation were evident on the response times of subjects, using a repeated measures analysis of variance (ANOVA) [F (1, 18) = 1.408, p = .251], and no significant interaction occurred between type and side of stimulation and angle of rotation [F (3, 54) = .466, p = .220]. The results of the reaction times are presented in Figure 2. To examine the effect of repetititve transcranial magnetic stimulation (rTMS) on accuracy of performance, a repeated-measures ANOVA with stimulation type (real, sham) × side (left, right) × angle (0°, 60°, 120°, 180°) as within-subject factors found a significant main effect of stimulation [F (1, 18) = 5.489, p = .031]. There was, as expected, a significant difference in
752
Eva A. Feredoes and Perminder S. Sachdev
Fig. 2 – Mean reaction times under each condition, which increased with angular disparity, except for inverted (180°) stimuli. The TMS conditions did not differ significantly from sham conditions. Error bars indicate ± 1 SEM.
accuracies between the angles of rotation [F (3, 54) = 56.968, p = .0001]. A significant interaction was evident between the type and side of stimulation and the angle of rotation [F (3, 54) = 3.110, p = .034]. Pairwise comparisons showed that right PPC TMS reduced accuracy for 60° rotation nonsignificantly (p = .087; Bonferroni corrected), and for 120° significantly (p = .003; Bonferroni corrected) but not for 180° rotation (Figure 3). Left-sided stimulation on the other hand, significantly reduced accuracy for 180° rotation (p = .012; Bonferroni corrected), but not for the other angles examined (Figure 4). DISCUSSION
Fig. 3 – Mean accuracy under right PPC TMS compared to sham. Stimulation significantly decreased accuracy for 120°rotated angles. Error bars indicate ± 1 SEM.
Fig. 4 – Mean accuracy under left PPC TMS compared to sham. Stimulation significantly decreased accuracy for 180°rotated angles. Error bars indicate ± 1 SEM.
The aims of this investigation were to confirm the contribution of the right PPC and explore possible involvement of the left hemisphere in the mental rotation of complex 3D stimuli. Harris and Miniussi (2003) have previously demonstrated that mental rotation was disrupted when TMS was delivered over the right PPC, 400-600 msec poststimulus onset. They hypothesised that the more complex Shepard and Metzler (1971) shapes may recruit the left PPC, and we set out to test this using TMS. Our results replicated the findings of Harris and Miniussi (2003) that right-sided TMS disrupted task performance, which was statistically significant for stimuli rotated by 120°. This is further evidence for the central role this region plays in the processing of spatial transformations required for mental rotation. Stimuli rotated by 120° demand a comparatively larger amount of mental processing, as reflected by the increased errors and response times under control conditions, and are therefore particularly susceptible to interference. TMS also reduced the accuracy of 60° rotation but this did not reach significance. TMSinduced performance deficits are more likely to occur when visual stimulus presentation is brief, or close to threshold of performance (Walsh and Pascual-Leone, 2003). In this study the duration of
Differential effects of rTMS on mental rotation
TMS was very short relative to the long stimulus viewing times. This is also likely to have masked any effects of TMS on reaction time because responses were not under time pressure (see Walsh and Pascual-Leone, 2003, for a full methodological discussion). It could also be argued that the effects produced by TMS can be explained by task difficulty, disrupting only the stimuli with the longest reaction times. This, however, is unlikely as left and right PPC TMS had different effects on stimuli that had similar reaction times. Interestingly, left PPC stimulation produced a significant decrease in accuracy for inverted stimuli (180° rotation). The result indicates a role for this region in mental rotation that is different from the role played by the right PPC. A closer examination of the special nature of 180° rotations may offer an explanation. This observation that not all angles of rotation are processed equally has been shown with respect to inverted stimuli (Turnbull et al., 2004; Walsh and Butler, 1996) where the increase in response times do not follow the expected linear pattern from 120° to 180°. Our behavioural results replicate this common finding. As vertically inverted stimuli are often simply an upside-down version of the target, Cooper and Shepard (1973) proposed that instead of piecemeal rotation used for smaller angles of rotation, subjects flip the whole object about the horizontal axis and then about the vertical (if mirrored; see also Walsh and Butler, 1996 and Carpenter et al., 1999) and response times are increased when a lateral plus vertical rotation must be made, compared to only vertical rotations. Different processing mechanisms between rotated and inverted shapes have also been reported in the monkey lesion literature and human neuropsychological literature. Parietal lesions in monkeys produced deficits in recognising shapes that had been left-right inverted but not for updown inversions (Ockleford et al., 1977). A left temporo-parietal lesion in a human subject showed the same pattern of results (Riddoch and Humphreys, 1988). It has been suggested that in the natural environment, left-right orientation is not important for discriminating between two objects, as this usually reflects the vertical symmetry natural to faces, front views of bodies and many living organisms (Gross and Bornstein, 1978). These enantiomorphic forms are therefore regarded by the brain as perceptually equivalent (Gross and Bornstein, 1978) and differentiating between two such shapes requires a greater degree of spatial processing (Walsh and Butler, 1996). Indeed, the unusual nature of enantiomorphs has been suggested as an explanation for the many cells in the inferotemporal cortex which respond to objects particularly strongly for their mirror inversions (Logothetis et al., 1995). According to Walsh and Butler (1996),
753
recognition of vertically inverted stimuli is carried out by areas more involved in object rather than spatial processing. This is supported by our results which show that under right-sided stimulation, the right side being pre-eminent in visuospatial processing; there is no effect of TMS on inverted stimuli specifically. Only left-sided stimulation was able to significantly disrupt these stimuli. The involvement of the left hemisphere in visual imagery is well-established, specifically for the verbal categorical encoding of an object (Kosslyn et al., 1995; Michimata, 1997; Parrot et al., 1999). Spatial relations can be represented by categories that can be easily verbalised, such as “above/below, left to/right to”, or alternatively, by a coordinate strategy that encodes the metric information of an object (Kosslyn et al., 1995). Neuroimaging studies have found preferential left parietal activation for categorical tasks, whilst the right is more active when precise spatial information is utilised (Baciu et al., 1999; Trojano et al., 2002). It is possible that inverted stimuli were processed using verbally-based categorical information, mediated by the left PPC and as a consequence, were the only such stimuli to be disrupted by left-sided stimulation. A role of the left parietal cortex in mental rotation has been difficult to establish. Alivisatos and Petrides (1997) found left parietal activity specific to mental rotation of alphanumeric characters, as did Vingerhoets et al. (2001) for paired 2D stimuli. Other studies have reported bilateral parietal involvement (Carpenter et al., 1999; Kosslyn et al., 1998; Richter et al., 1997; Tagaris et al., 1997) or only right hemispheric processing (Deutsch et al., 1988; Harris et al., 2000). These differences could be due to procedural differences between investigations, such as the different range of stimuli orientations, instructions given, task practice, or the type of neuroimaging approach used (Harris et al., 2000). Corballis (1997) suggested left-hemisphere involvement in mental rotation tasks utilising more complex stimuli due to subjects’ use of verbal or sequential strategies. Neuro- and experimental psychology evidence is available to support this (Corballis, 1997; McGuinness and Bartell, 1982; Mehta and Newcombe, 1991) and subjective reports from normal subjects are that verbal or symbolic strategies are used when rotating complex 3D (but not 2D) figures (McGuinness and Bartell, 1982). Monkey lesion evidence is also available showing that stimuli of different dimensions are not processed in the same way (Holmes and Gross, 1984). CONCLUSION Our investigation supports the important role of right PPC in mental rotation. It also shows that the left PPC plays an important and hitherto under-
754
Eva A. Feredoes and Perminder S. Sachdev
investigated role in mental rotation, especially when it involves inversion of a 3D figure. Acknowledgements. We are incredibly grateful to Vincent Walsh for his expert comments on the manuscript and to Garth Hargreaves for developing the computerised stimuli. REFERENCES ALIVISATOS B and PETRIDES M. Functional activation of the human brain during mental rotation. Neuropsychologia, 35: 111-118, 1997. BACIU M, KOENIG O, VERNIER MP, BEDOIN N, RUBIN C and SEGEBARTH C. Categorical and coordinate spatial relations: fMRI evidence for hemispheric specialization. Neuroreport, 10: 1373-1378, 1999. CARPENTER PA, JUST MA, KELLER TA and EDDY W. Graded functional activation in the visuospatial system with the amount of task demand. Journal of Cognitive Neuroscience, 11: 9-24, 1999. COHEN MS, KOSSLYN SM, BREITER HC, DIGIROLAMO GJ, THOMPSON WL, ANDERSON AK, BOOKHEIMER SY, ROSEN BR and BENNETT J. Changes in cortical activity during mental rotation: A mapping study using functional MRI. Brain, 119: 89-100, 1996. COOPER LA and SHEPARD RN. Chronometric studies of the rotation of mental images. In Chase WG (Ed), Visual Information Processing. New York: Academic Press, 1973. CORBALLIS MC. Mental rotation and the right hemisphere. Brain and Language, 57: 100-121, 1997. DEUTSCH G, BOURBON WT, PAPANICOLAOU AC and EISENBERG HM. Visuospatial tasks compared via activation of regional cerebral blood flow. Neuropsychologia, 26: 445-452, 1988. GROSS CG and BORNSTEIN MH. Left and right in science and art. Leonardo, 11: 29-38, 1978. HARRIS IM, EGAN GF, SONKKILA C, TOCHON-DANGUY HJ, PAXINOS G and WATSON JDG. Selective right parietal lobe activation during mental rotation: A parametric PET study. Brain, 123: 65-73, 2000. HARRIS IM and MINIUSSI C. Parietal lobe contribution to mental rotation demonstrated with rTMS. Journal of Cognitive Neuroscience, 15: 315-323, 2003. HOLMES EJ and GROSS CG. Stimulus equivalence after inferior temporal lesions in monkeys. Behavioral Neuroscience, 98: 898-901, 1984. KOSSLYN SM, DIGIROLAMO GJ, THOMPSON WL and ALPERT NM. Mental rotation of objects versus hands: Neural mechanisms revealed by positron emission tomography. Psychophysiology, 35: 151-161, 1998. KOSSLYN SM, MALJKOVIC V, HAMILTON SE, HORWITZ G and THOMPSON WL. Two types of image generation: Evidence for left and right hemispheric processes. Neuropsychologia, 33: 1485-1510, 1995.
LOGOTHETIS NK, PAULS J and POGGIO T. Shape representation in the inferior temporal cortex of monkeys. Current Biology, 5: 552-563, 1995. MCGUINNESS D and BARTELL DE. Hard or simple-minded? Neuropsychologia, 20: 629-639, 1982. MEHTA Z and NEWCOMBE F. A role for the left hemisphere in spatial processing. Cortex, 23: 163-167, 1991. MICHIMATA C. Hemispheric processing of categorical and coordinate spatial relations in vision and visual imagery. Brain and Cognition, 33: 370-387, 1997. OCKLEFORD EM, MILNER AD, DEWAR W and SNEDDON IA. Form perception in stumptail macaques following posterior parietal and lateral frontal lesions. Cortex, 13: 361-372, 1977. PARROT M, DOYON B, DEMONET JF and CARDEBAT D. Hemispheric preponderance in categorical and coordinate visual processes. Neuropsychologia, 37: 1215-1225, 1999. PEGNA AJ, KHATEB A, SPINELLI L, SEECK M, LANDIS T and MICHEL CM. Unraveling the cerebral dynamics of mental imagery. Human Brain Mapping, 5: 410-421, 1997. RATCLIFF G. Spatial thought, mental rotation and the right cerebral hemisphere. Neuropsychologia, 17: 49-54, 1979. RICHTER W, UGURBIL K, GEORGOPOLOUS AP and KIM SG. Timeresolved fMRI of mental rotation. Neuroreport, 8: 3697-3702, 1997. RIDDOCH MJ and HUMPHREYS GW. Description of a left/right coding deficit in a case of constructional apraxia. Cognitive Neuropsychology, 5: 289-315, 1988. SHEPARD RN and METZLER J. Mental rotation of three-dimensional objects. Science, 19: 701-703, 1971. TAGARIS GA, KIM SG, STRUPP JP, ANDERSON P, UGURBIL K and GEORGOPOLOUS AP. Mental rotation studied by functional magnetic resonance imaging at high field (4T): Performance and cortical activation. Journal of Cognitive Neuroscience, 9: 419-432, 1997. TROJANO L, GROSSI D, LINDEN DE, FORMISANO E, GOEBEL R, CIRILLO S, ELEFANTE R and DI SALLE F. Coordinate and categorical judgments in spatial imagery: An fMRI study. Neuropsychologia, 40: 1666-1674, 2002. TURNBULL OH, DRIVER J and MCCARTHY RA. 2D but not 3D: Pictorial-depth deficits in a case of visual agnosia. Cortex, 40: 723-738, 2004. WALSH V and BUTLER SR. The effects of visual cortex lesions on the perception of rotated shapes. Behavioural Brain Research, 76: 127-142, 1996. WALSH V and PASCUAL-LEONE A. Transcranial magnetic stimulation: A neurochronometrics of mind. Cambridge, MA: MIT Press, 2003. W ASSERMAN E. Risk and safety of repetitive transcranial magnetic stimulation: report and suggested guidelines from the International Workshop in the Safety of Repetitive Transcranial Magnetic Stimulation. Electroencephalography and Clinical Neurophysiology, 108: 1-16, 1998.
Eva A. Feredoes, 1202 West Johnson Street, Madison WI 53706, USA. e-mail: [email protected]