Cross-adaptation combined with TMS reveals a functional overlap between vision and imagery in the early visual cortex
Introduction
Visual imagery, a process subserved by the visuospatial working memory system (see e.g. Baddeley, 2003), refers to the generation of an internal pictorial representation following a description of the target stimulus (Baddeley, 1986, Cornoldi and Vecchi, 2003). On the neural level, there is evidence that working memory processes such as imagery and visual short-term memory are mediated by the same cortical areas that process incoming sensory information (cf. Barsalou, 2008, Kosslyn and Thompson, 2003; see also the “sensory–recruitment” model of WM, see e.g. Awh and Jonides, 2001, D'Esposito, 2007). In the context of visual imagery, a number of neuroimaging studies support this view, implicating a wide range of visual cortical areas in visual imagery (e.g. Ganis et al., 2004, Kaas et al., 2010, Klein et al., 2000, Kosslyn et al., 1999, Seurinck et al., 2011, Slotnick et al., 2005). However, the role of the early visual areas in visual mental imagery is not fully clear. While V1 has been implicated by some studies (e.g. Ganis et al., 2004, Kosslyn et al., 1999, Slotnick et al., 2005), others have found no involvement of this region in visual imagery (Ghaem et al., 1997, Kaas et al., 2010, Knauff et al., 2000, Roland and Gulyas, 1995, Wheeler et al., 2000). For example, Kaas et al. (2010) recently found that activation of extrastriate areas was associated with the content of visual imagery, whereas the early visual cortex was down-regulated during imagery.
Transcranial magnetic stimulation (TMS) studies which have tested the causal role of the early visual cortex in mental imagery have found TMS to impair performance in visual imagery tasks (Cattaneo et al., 2009a, Kosslyn et al., 1999). However, it is not clear whether these TMS effects reflect the disruption of the content of mental imagery, or whether they result from the disruption of processes which are only indirectly related to the mental image (e.g. attention). In order to directly assess the functional overlap in neuronal representations between visual perception and mental imagery, we combined TMS with an adaptation paradigm. The advantage in using an adaptation paradigm is that it enables one to demonstrate an overlap in the neuronal representations of two stimuli: for example, if adaptation to Stimulus 1 affects the subsequent encoding of Stimulus 2, this indicates the existence of a neuronal representation which encodes both stimuli (see e.g. Grill-Spector et al., 2006). Behaviorally, adaptation is reflected as a behavioral impairment on trials when the adapter and the test stimulus are encoded by the same neuronal representation relative to trials when the adapter and test do not share the same representation.
After visual adaptation, online TMS facilitates the detection of the adapted feature, if the targeted region contains neuronal representations that play a causal role in the encoding of that feature (Cattaneo et al., 2008, Cattaneo et al., 2009a, Cattaneo et al., 2009b, Cattaneo et al., 2010, Cattaneo et al., 2011, Cohen Kadosh et al., 2010, Silvanto et al., 2007). This selective facilitation of adapted attributes enables the combination of adaptation and TMS to reveal a functional overlap in neuronal representations (Silvanto and Muggleton, 2008). We hypothesized that if the same neuronal representations in the early visual cortex are engaged in both visual perception and imagery, then TMS should facilitate visual imagery when the mental image has to be generated in the adapted region of visual space. In contrast, if the role of the early visual cortex is limited to processes which are only indirectly linked to the content of the mental image (such as attention), then the impact of TMS should not depend on the state of adaptation.
Visual mental imagery was assessed by using a task derived from the mental clock test originally devised by Paivio (1978). In this version of the clock test (firstly used in a previous TMS study by Cattaneo et al., 2009a), participants are presented with a digital time (e.g. “2.15”) and asked to generate a mental image of the clock hands associated with this time. In the study by Cattaneo et al. (2009a), TMS applied over the early visual cortex impaired imagery performance and hence the same stimulation site was used here. The accuracy of the mental image was assessed by asking subjects to discriminate the position of a visually presented dot in relation to the imagined clock hands. In Experiment 1, we investigated whether visual adaptation to oriented lines presented either below or above fixation affects subsequent performance in the clock task (see Fig. 1).
The results of Experiment 1 show that visual imagery performance is worse when the mental image is generated in the adapted region of space relative to a nonadpated region of space. The presence of a behavioral cost of adaptation indicates that the paradigm was effective. In Experiment 2, we investigated whether this adaptation effect occurs in the early retinotopic visual areas by applying TMS over the occipital pole.
Section snippets
Methods
All experiments were undertaken with the understanding and written consent of each participant. The study had been approved by the institutional review board and all subjects were treated in accordance with the Declaration of Helsinki. Two experiments were completed as follows.
Experiment 1
Trials were classified according to whether the mental image spatially overlapped with the location of the adapter. There were hence three possible conditions: the “Spatial Overlap”, the “No Spatial Overlap”, and the “No adaptation” conditions. Fig. 2 shows participants' mean accuracy (n = 13) in the imagery task as a function of the spatial overlap between the mental image location and the adapter. In Experiment 1, a one-way ANOVA on participants' accuracy with the experimental condition
Discussion
Our behavioral results show that mental imagery accuracy was impaired when the spatial location in which the mental image had to be generated had been adapted by visual stimulation. This pattern of results was found both in Experiment 1 and in the No TMS and Control (Vertex TMS) conditions of Experiment 2. That a visually induced suppression of neuronal representations encoding a specific region of visual space affects mental image generation in that spatial location (but not in other
Acknowledgments
ZC has been supported by Banca del Monte di Pavia. JS is supported by the Academy of Finland.
References (46)
- et al.
Overlapping mechanisms of attention and spatial working memory
Trends Cogn. Sci.
(2001) Working Memory
(1986)Working memory: looking back and looking forward
Nat. Rev. Neurosci.
(2003)Grounded cognition
Annu. Rev. Psychol.
(2008)The relationship between visual perception and visual mental imagery: a reappraisal of the neuropsychological evidence
Cortex
(2002)- et al.
Priming of motion direction and area V5/MT: a test of perceptual memory
Cereb. Cortex
(2002) - et al.
Investigating visual motion perception using the transcranial magnetic stimulation–adaptation paradigm
Neuroreport
(2008) - et al.
Using state-dependency of transcranial magnetic stimulation (TMS) to investigate letter selectivity in the left posterior parietal cortex: a comparison of TMS-priming and TMS-adaptation paradigms
Eur. J. Neurosci.
(2008) - et al.
Contrasting early visual cortical activation states causally involved in visual imagery and short-term memory
Eur. J. Neurosci.
(2009) - et al.
TMS-adaptation reveals abstract letter selectivity in the left posterior parietal cortex
Cereb. Cortex
(2009)
State-dependent TMS reveals a hierarchical representation of observed acts in the temporal, parietal, and premotor cortices
Cereb. Cortex
One's motor performance predictably modulates the understanding of others' actions through adaptation of premotor visuo-motor neurons
Soc. Cogn. Affect. Neurosci.
Double dissociation of format-dependent and number-specific neurons in human parietal cortex
Cereb. Cortex
Visuo-spatial working memory and individual differences
From cognitive to neural models of working memory
Philos. Trans. R. Soc. Lond. B Biol. Sci.
Brain areas underlying visual mental imagery and visual perception: an fMRI study
Brain Res. Cogn. Brain Res.
Mental navigation along memorized routes activates the hippocampus, precuneus, and insula
Neuroreport
Neural adaptation of imaginary visual motion
Cogn. Psychol.
The constructive nature of vision: direct evidence from functional magnetic resonance imaging studies of apparent motion and motion imagery
Eur. J. Neurosci.
Repetition and the brain: neural models of stimulus-specific effects
Trends Cogn. Sci.
Decoding reveals the contents of visual working memory in early visual areas
Nature
Cortical interactions in vision and awareness: hierarchies in reverse
Prog. Brain Res.
Imagery of a moving object: the role of occipital cortex and human MT/V5 +
NeuroImage
Cited by (24)
Not all visual symmetry is equal: Partially distinct neural bases for vertical and horizontal symmetry
2017, NeuropsychologiaCitation Excerpt :Comparing the critical time window of OFA in symmetry detection with that of face processing (e.g., Pitcher et al., 2007), would allow to clarify the interaction between symmetry and face detection in this region. Moreover, future research may further disentangle the role of OFA in symmetry and face detection by employing priming or adaptation paradigms (e.g. Cattaneo and Silvanto, 2008; Cattaneo et al., 2012): this would allow to examine whether the same neural populations in OFA are indeed responsible for face and symmetry detection. In sum, our results shed light on the cortical underpinnings of vertical and horizontal symmetry encoding and demonstrate that the two symmetry orientations rely on partially different neural networks: in particular, we revealed that while right LO plays a causal role in perception of both symmetry orientations, the role of right OFA is limited to vertical symmetry and does not extend to horizontal orientation.
Can predictive coding explain repetition suppression?
2016, CortexCitation Excerpt :The actual sensory input on the other hand travels through the system in the opposite direction, with first lower-level ε calculations possibly occurring as early as the LGN. After all, since there is evidence for both RS effects and PC in the early visual cortex as well as the LGN (Cattaneo, Bona, & Silvanto, 2012; Jehee & Ballard, 2009; Larsson & Smith, 2012; Ortuno, Grieve, Cao, Cudeiro, & Rivadulla, 2014; Rauss, Schwartz, & Pourtois, 2011), it is likely that these areas contribute to the low-level prediction error computations as well. We would like to emphasize however, that this represents a rather simplified model of the underlying network, as most likely all depicted regions (apart from the LGN) are interconnected to some degree.
Non-invasive Human Brain Stimulation in Cognitive Neuroscience: A Primer
2015, NeuronCitation Excerpt :Subsequent uses of the state-dependent paradigm have proven its utility in several domains. Cattaneo et al., (2012) showed that adapting a region of the visual field led to impairments in mental imagery in that region of space and that this inhibition was unmasked by the application of TMS to occipital visual cortex. Moving up the processing hierarchy, Silvanto and Soto (2012) used state dependency to show that TMS facilitated performance on a visual short-term memory task.
Partial dissociation in the neural bases of VSTM and imagery in the early visual cortex
2015, NeuropsychologiaMental Imagery: Visual Cognition
2015, International Encyclopedia of the Social & Behavioral Sciences: Second Edition