Individual Differences in Mental Imagery Modulate Effective Connectivity of Scene-Selective Regions During Resting State

doi:10.21203/rs.3.rs-887061/v1

Download PDF

Research Article

Individual Differences in Mental Imagery Modulate Effective Connectivity of Scene-Selective Regions During Resting State

https://doi.org/10.21203/rs.3.rs-887061/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 21 Mar, 2022

Read the published version in Brain Structure and Function →

You are reading this latest preprint version

Successful navigation relies on the ability to identify, perceive, and correctly process the spatial structure of a scene. It is well known that visual mental imagery plays a crucial role in navigation. Indeed, cortical regions encoding navigationally relevant information are also active during mental imagery of navigational scenes. However, it remains unknown whether their intrinsic activity and connectivity reflect the individuals’ ability to imagine a scene. Here, we primarily investigated the intrinsic causal interactions among scene-selective brain regions such as Parahipoccampal Place Area (PPA), Retrosplenial Complex (RSC), and Occipital Place Area (OPA) using Dynamic Causal Modelling (DCM) for resting-state functional magnetic resonance (rs-fMRI) data. Second, we tested whether resting-state effective connectivity parameters among scene-selective regions could reflect individual differences in mental imagery in our sample, as assessed by the self-reported Vividness of Visual Imagery Questionnaire (VVIQ). We found an inhibitory influence of occipito-medial on temporal regions, and an excitatory influence of more anterior on more medial and posterior brain regions. Moreover, we found that a key role in imagery is played by the connection strength from OPA to PPA, especially in the left hemisphere, since the influence of the signal between these scene-selective regions positively correlated with good mental imagery ability. Our investigation contributes to the understanding of the complexity of the causal interaction among brain regions involved in navigation and provides new insight in understanding how an essential ability, such as mental imagery, can be explained by the intrinsic fluctuation of brain signal.

Cellular & Molecular Neuroscience

scene-selective regions

navigation

imagery

fMRI

resting-state

dynamic causal modelling

To explore the surrounding environment our brain must be able to recognize, memorize and recall specific places in the world. More specifically, forming and memorizing a mental image (i.e., a cognitive map) of a specific place plays a crucial role in our ability to navigate the spatial world. For this reason, it is plausible that being a good imager might enhance the skills to navigate in the environment. Over the past two decades, some important issues have been clarified regarding the link between imagery and navigational performance. Recently, functional resonance imaging (fMRI) studies clarified that voluntary mental imagery activates several brain regions spanning from frontal areas, such as the dorsal and ventral prefrontal cortex, to medial and temporal areas such as the hippocampus and category-selective regions (Mechelli, 2004; Ranganath and D’Esposito, 2005; Pearson and Westbrook, 2015). A top-down model has been introduced and proposes that mental images are organized and controlled by frontal regions including the inferior frontal gyrus and anterior cingulate cortex (Fulford et al., 2018), information stored in memory is retrieved in medial temporal areas such as the hippocampus, and the content-specific representation of the mental images activates regions in the most posterior part of the brain (Dijkstra et al., 2017). Indeed, the specific content of a mental image activates different brain regions, similarly to what happens during visual perception (Ganis et al., 2004). For instance, imagining a face leads to activation of the fusiform face area (FFA), a brain area activated during face perception; similarly, imaging a scene leads to activation of ventromedial posterior cortical regions (scene-selective regions), which are known to be activated during scene and landscape perception (Epstein and Baker, 2019).

The scene-selective regions include at least three visual cortical areas responding selectively to scenes, compared to other category images such as faces and objects. Specifically, these regions are termed the Parahippocampal Place Area (PPA), the Retrosplenial Complex (RSC), and the Occipital Place Area (OPA), which are temporal, medial, and occipital areas respectively. These regions play different and complementary roles, and several neuroimaging studies demonstrated the contribution of each aforementioned brain region in scene processing. PPA, located at the boundary between the posterior parahippocampal cortex and the anterior lingual gyrus, is activated by passive viewing of real-world scenes or landmarks (Epstein and Kanwisher, 1998); The Retrosplenial complex extends from the retrosplenial cortex itself, located posteriorly to the corpus callosum, through the posterior cingulate cortex to the anterior bank of the parietal-occipital fissure (Burles et al., 2017). RSC is mostly active during both real and imagined navigation (Maguire, 2001; Ino et al., 2002; Wolbers, 2005), retrieval of environment-centered information (Committeri et al., 2004; Galati et al., 2010), and is selectively sensitive to a change in point of view (Sulpizio et al., 2016), mental imagery of familiar places (Boccia et al., 2015) and encoding of permanent items (Auger et al., 2012; Auger and Maguire, 2013). More recently, a growing number of studies have disclosed the role of OPA in scene perception. OPA, which is located around the transverse occipital sulcus is retinotopically organized (Nasr et al., 2011) and shows a preference for the lower visual field (Silson et al., 2015), encodes environmental boundaries (Julian et al., 2016) and local navigational affordances (Bonner and Epstein, 2017), and represents motion information in the immediately visible scene from a first-person perspective (Kamps et al., 2016). These results suggest that OPA is specialized in encoding both low and high-level characteristics such as environmental affordances of a scene (Epstein and Baker, 2019) and might be a primary site toward higher cortical regions.

Several neuroimaging human studies revealed that PPA, RSC, and OPA are strongly interconnected. Studying functional connectivity among scene-selective and hippocampal regions at rest, Boccia et al. (Boccia et al., 2017a) found that PPA is connected to occipito-temporal areas including RSC, lingual gyrus, the calcarine cortex, the parieto-occipital sulcus, and the posterior hippocampus. Similarly, Silson et al. (2015) found that OPA is significantly more connected to the posterior portion of the parahippocampal area (pPPA) than to the anterior portion of the parahippocampal area (aPPA) while the retrosplenial complex/medial parietal cortex (MPA) is more strongly connected with aPPA than pPPA. Similar results were obtained by Baldassano et al. (2013)

In summary, PPA, RSC, and OPA are key nodes of a network supporting both perception and imagery of environmental scenes. Although the functional coupling between these regions seems to be well established in humans based on previous resting-state functional imaging studies (Margulies et al., 2009; Nasr et al., 2013), information about the directed causal interactions among PPA, RSC, and OPA is still lacking. Here, we used spectral Dynamic Causal Modelling (DCM) (Friston et al., 2003, 2014) for resting-state fMRI to assess the strength of the influence that these regions exert on each other without any explicit task. After assessing the dynamic functional connectivity in terms of directed effective connectivity among scene-selective regions, we also tested whether individual differences in connectivity estimates could be related to individual differences in mental imagery ability.

By using resting-state functional connectivity, we previously found that the pattern of reciprocal connections between scene-selective regions reflects individual differences in spatial navigation (Sulpizio et al., 2016; Tullo et al., 2018). In particular, the vividness of visual imagery seems to have a key role in a wide range of cognitive abilities (Pearson, 2019), including successful performance on mental rotation (Pazzaglia and Moè, 2013), spatial tasks (Piccardi et al., 2017), and navigation (Marchette et al., 2011; Keogh and Pearson, 2011, 2014). “Seeing with the mind’s eye”, i.e., imaging something, assumes that visual information arises from memory or likewise can be combined and modified to make a strategy or to achieve a goal (Kosslyn et al., 2001). In this sense, having a good imagery ability plays a crucial role in constructing and recalling a schematic representation of the environment (e.g., a map), adapting own position to the surrounding place, and consequently getting the easiest way-finding. In this framework, several studies revealed that individual differences in cognitive style such as visual or verbal strategy (Blazhenkova and Kozhevnikov, 2009, 2010) affect the type of information individuals choose to guide navigation (Pazzaglia and Moè, 2013; Kraemer et al., 2017; Piccardi et al., 2017). Indeed, subjects who select and search for salient landmarks rely on a landmark strategy; subjects who learn the path using egocentric coordinates (for example where to turn, right or left) rely on a route strategy; subjects who use both egocentric and allocentric coordinates rely on a survey strategy and they have a clearer global map of the surrounding environment (Pazzaglia and De Beni, 2001). It has been found that survey individuals re-orient themselves faster than landmarks and route-individuals because they are more independent from the space around them. Mental imagery is at the base of this ability. It has been found that a preference for verbalizing descriptions is useful for retrieving position information but not for giving the relative spatial location, while a visual cognitive strategy is useful for judging relative direction among landmarks (Nori and Giusberti, 2006; Kraemer et al., 2017).

Here we used DCM applied to rs-fMRI and a self-report questionnaire on mental imagery skills, which assess the vividness of visual mental images (Vividness of Visual Imagery Questionnaire, VVIQ) (Marks, 1973), to test the hypothesis that individual differences in mental imagery abilities account for how scene-selective regions interact with each other. Results of this study point out that causal information flow among scene-selective regions is relevant to understand individual differences, in particular mental imagery ability that plays a crucial role in successful navigation in the surrounding environment.

Participants

A total of forty-two healthy volunteers (mean age = 32.21, SD = 4.17, 24 female) with normal or corrected vision participated in this study. Each subject underwent two fMRI sessions including two localizer scans for scene-selective regions (4’24’’ each) and two scans of resting-state (8’18’’ each). Each scan began less than a minute after the other unless the subject asked for a brief break inside the scanner. Magnetic resonance imaging (MRI) data were collected on a Siemens Allegra 3T at the Santa Lucia Foundation in Rome (Italy) for 18 of the 42 participants and on a Philips Achieva 3T scanner at the Institute for Advanced Biomedical Technologies (ITAB) of the University G. D’Annunzio Foundation in Chieti (Italy) for remaining 24 subjects. All participants were right-handed, as assessed by the Edinburgh Handedness Inventory (Oldfield, 1971) and had a normal or corrected-to-normal vision. All volunteers gave their written informed consent to participate in the study that was previously approved by the research ethics committees either at Fondazione Santa Lucia or at University G. D’Annunzio, according to the Declaration of Helsinki.

Experimental paradigm

The localizer fMRI experiment consisted of eight blocks (16 seconds) of passive viewing of faces alternated with eight blocks (16 seconds) of passive view of navigationally relevant stationary stimuli such as places, with each image presented for 300 ms every 500 ms, interleaved with fixation periods of 15 seconds on average. Half of the pictures consisted of common indoor and half of common outdoor scenes. Pictures of faces represented faces with neutral expressions of male (50%) and female (50%) young adults. In the resting-state fMRI experiments, subjects were laying with eyes closed, they were asked not to think about anything in particular, and no experimental task was imposed.

Vividness of Visual Imagery Questionnaire

The vividness of visual imagery was measured using the Vividness of Visual Imagery Questionnaire (VVIQ) (Marks, 1973). The VVIQ is a self-reported questionnaire composed of 16 items arranged in blocks of four items. Subjects were asked to visualize specific images (e.g., “visualize a raising sun. Carefully consider the following picture that comes in your mind’s eye: a rainbow appears”), and to rate the vividness of each image using a 5-point Likert scale from 1 (clear image such as a real view) to 5 (no image). Twenty-eight of forty-two participants answered to VVIQ questionnaire. We reversed participants’ scores to have low scores for poorer imagery ability and high scores for better imagery ability.

Image and acquisition processing

Functional T2*-weighted images were collected using a gradient-echo EPI sequence using blood-oxygenation level-dependent (BOLD) contrast over the whole brain (Kwong et al., 1992). BOLD scans collected on Siemens Allegra 3T scanner included thirty contiguous 4-mm slices and were acquired with an in-plane resolution of 3 x 3 mm in an interleaved excitation order (echo time [TE] = 30 ms, repetition time [TR] = 2 s, flip angle = 70°). For localizer and resting-state scans, 242 and 128 volumes were respectively acquired. BOLD images collected on Philips Achieva 3T scanner included thirty-nine contiguous 4-mm slices, acquired in an interleaved order (voxel size = 3.6 x 3.6 x 3.6 mm, [TE] = 25 ms, [TR] = 1.914 s, flip angle = 80°). For localizer and resting-state scans 249 and 160 volumes were respectively acquired. Structural images were collected using a sagittal magnetization-prepared rapid acquisition gradient echo (MPRAGE) T1-weighted sequence. Imaging parameters for Allegra MPRAGE scans were as follows: [TE] = 4.4 ms, flip angle = 8°, in-plane resolution = 0.5 x 0.5 mm, slice thickness = 1 mm. Imaging parameters for Achieva MPRAGE scans were as follows: [TE] = 3.7 ms, flip angle = 8°, voxel size = 1 x 1 x 1 mm.

Preprocessing was performed using the SPM12 software package (Wellcome centre for Human Neuroimaging, London). In each scan, we discarded the first four volumes from data analysis to exclude non-steady-state scans. Resting-state and localizer images were corrected for differences in slice timing, using the central slice of each volume as a reference. Images were realigned to the first functional volume of each session and were coregistered to the skull-stripped anatomical image. Finally, images were normalized to MNI space (Mazziotta et al., 1995) and smoothed using a Gaussian kernel with a 6-mm FWHM. A resting-state effective connectivity analysis between theoretically motivated regions of interest (ROIs) was then performed.

Region of Interest definition

Regions of interest were defined in volume space on each individual hemisphere separately: Parahippocampal Place Area (PPA), Retrosplenial complex (RSC), and Occipital Place Area (OPA) were defined in each participant by analyzing localizer imaging runs. Place and face blocks were modelled with box-car functions, convolved with a canonical hemodynamic response function. The scene-responsive ROIs were defined as the regions responding more strongly to places than to faces. Each ROI was defined as the conjunction of the results of a T-contrast (places > faces), a sphere (8mm radius) around the subject-specific local maxima, and a 10 mm sphere radius around ROI centre coordinates defined at the group level. The statistical parametric map resulting was thresholded at p < 0.05 FDR corrected at the cluster level, after applying a cluster-forming threshold of p < 0.001 uncorrected at the voxel level. The anatomical localization of ROIs is shown in Fig. 1, whereas the mean coordinates and standard deviation of the regional peaks are detailed in Table 1.

Table 1

Mean and Standard Deviation of MNI coordinates (X, Y, Z) of the Parahippocampal Place Area (PPA), the Occipital Place Area (OPA) and the Retrosplenial Complex (RSC) in left and right hemispheres.
			MNI Coordinates
Hemisphere	Regions	X	Y	Z
Left	PPA	-24.5 ± 2.1	-47.4 ± 2.2	-11.9 ± 2.3
	OPA	-31.0 ± 2.1	-85.7 ± 2.8	16.6 ± 3.3
	RSC	-20.0 ± 2.2	-61.5 ± 3	14.0 ± 3.7
Right	PPA	27.4 ± 1.9	-50.7 ± 3.7	-10.5 ± 2.6
	OPA	34.7 ± 2.6	-82.4 ± 2.3	12.1 ± 3.1
	RSC	21.1 ± 1.9	-59.4 ± 2.7	14.7 ± 3.8

Resting-state data were first modelled with a general linear model (GLM) containing six head motion regressors (three translational, three rotational), cerebrospinal fluid (CSF) signal, white matter (WM) signal, and discrete cosine basis regressors (frequency range: 0.0078-0.1 Hz). An F-contrast was specified across all DCT components to produce an SPM. Time-series were extracted by computing the principal eigenvariate of signals from voxels centred on the peak voxel within each ROI and were subsequently used for DCM analyses.

Dynamic Causal Modelling (DCM) Analysis

We used spectral DCM for resting-state fMRI (Friston et al., 2014) to estimate intrinsic effective connectivity among the scene-selective brain regions from resting-state BOLD responses. Briefly, in the A-matrix we modelled the strength in units of Hertz (Hz) of the connection among regions. In other words, we observed the rate of change in activity (per second) in one region affected by another. As a constraint of our model, we used a fully connected model, i.e., each region was connected with the other two areas of the same hemisphere (Fig. 1). Each parameter of the DCM A-matrix has a biophysical interpretation since a positive value means that the connection is excitatory (i.e., a region increases the activity in another region); on the contrary, if the parameter has a negative value the connection is inhibitory (Zeidman et al., 2019). In line with what has been done before (Almgren et al. 2018), we specified and inverted DCMs (with all possible connections among regions, i.e., a “full model”) for each session separately without specifying any exogenous input (DCM 12; revision 6801). We excluded from the analysis the sessions that didn’t meet the following criteria: (1) explained variance of predicted BOLD signals above 60%, (2) at least one connection with a strength greater than 1/8 Hz, and (3) at least one effectively estimated parameter (based on Kullback-Leibler divergence of posterior from prior distribution). These criteria were used as diagnostic tools for detecting so-called ‘bad’ model fits. Since DCM uses a gradient-based approach to maximize the approximate log model evidence, the initial parameter regime might be located in a local plateau. As a consequence, the inversion scheme is ‘stuck’ and hence is unable to obtain a good fit with a valid log model evidence bound.

To assess the relationship between imaginative abilities and connectivity estimates, a multilevel hierarchical linear model using Parametric Empirical Bayes (PEB) was performed (Friston et al., 2016). At the subject level, the average connectivity across sessions was modelled using PEB. Then, two PEB analyses were performed: firstly, we performed a group-level PEB with a constant term (i.e., the average across subjects) for all the forty-two participants. Secondly, we performed a PEB using both a constant term and VVIQ-scores as regressors for respectively modelling the average connectivity across subjects and the individual differences in connectivity in terms of differences in imaging abilities in twenty-eight participants. A posterior probability criterion of 95% for each group-level parameter estimate was used to infer average effective connectivity and the relation with VVIQ.

Group average connectivity strengths

The PEB analysis, modelling the average connectivity on forty-two subjects, revealed that the left and the right hemispheres have similar intrinsic connectivity strength at rest (Fig. 2). Indeed, the Retroplenial complex and the Occipital Place area have an inhibitory role on the other regions in both hemispheres. Notwithstanding, the influence of right RSC on right OPA didn’t exceed the posterior probability (pp < 0.95). At the same time, the Parahippocampal Place area has an excitatory role onto the other regions more in the left than in the right hemisphere. However, the connectivity parameter from right PPA to right RSC didn’t survive to the posterior probability criterion in the right hemisphere.

Connectivity strengths and the relationship with VVIQ score

The vividness of Visual Imagery Questionnaire (VVIQ) was used to assess the vividness of mental images in twenty-eight of forty-two participants. Lower VVIQ scores reflected poorer imagery abilities and higher scores reflected better imagery abilities. The effect of VVIQ on each connectivity parameter is shown in Fig. 3. Concerning the VVIQ, a different association of the right and left hemispheres with visual imagery was found. In the left hemisphere, a negative association between imagery abilities and the outgoing connectivity from RSC to OPA (connection strength = -0.008, pp = 0.98) and from PPA to OPA (connection strength = -0.010, pp = 0.96) was found. On the contrary, the outgoing connectivity from left OPA to both the other scene-selective regions was positively related to good imagery abilities: the outgoing connectivity from OPA to PPA (connection strength = 0.011, pp = 0.94) and from OPA to RSC (connection strength = 0.008, pp = 0.85) was positively associated with VVIQ scores, even if the connectivity parameter between OPA and RSC didn’t exceed the threshold (pp = 0.85). Conversely, the posterior parameter from left OPA and left PPA (connection strength = 0.011) is equal to 0.94, an acceptable threshold value. In the right hemisphere, the connectivity from OPA to PPA is negatively related to good imagery (parameter = -0.041, pp = 1.00). Here, we didn’t consider the relationship between VVIQ and the self-connectivity (right PPA to right PPA = -0.024, pp = 1.00; right OPA to right OPA = 0.024, pp = 1.00) since we were interested in the modulatory connectivity among scene-selective regions.

In the present study, we primarily aimed at investigating the intrinsic causal interactions between scene-selective regions - the Parahippocampal Place Area (PPA), the Occipital Place Area (OPA) and the Retrosplenial Complex (RSC). Secondly, in a sub-sample we tested whether resting-state effective connectivity parameters were associated with mental imagery ability, as assessed by a self-reported questionnaire on vividness imagery. To that aim, we explored the causal influence of three regions involved in scene perception and imagery (PPA, RSC, OPA) in both hemispheres by using Dynamic Causal Modelling for resting-state fMRI. The first finding of this study is that both RSC and OPA have an inhibitory role at rest on the other regions and the more ventral region PPA excites the visual area OPA in the right hemisphere, while both OPA and RSC in the left hemisphere show an excitation feedback influence of higher-level regions on the lowest ones. It’s important to remember that we took into account the spontaneous connectivity among three scene-selective regions. Results were found to reflect the signal at baseline that could change depending on the task performed or the regions that may be included in the model. Since to our knowledge, there are no studies that have explored the direction of signal among the scene-selective regions at rest, these results give a first exploratory outlook on how these regions interact with each other.

In our results, OPA has an inhibitory influence on the other regions at rest. In visual stimuli perception, the primary visual areas excite the more anterior regions responding to the specific content of the image categorization (bottom-up). In other tasks, for example during imagery, the imagery input comes from the frontal regions towards the more posterior one (top-down) (Mechelli, 2004; Dijkstra et al., 2017). It is plausible to hypothesize that during a visual perception task the connectivity strength may be excitatory rather than inhibitory such as during rest. At the same time, other regions such as RSC are part of the “default mode network” (DMN) in which the activity at baseline is always higher than in other brain regions and their functions are never turned off but only enhanced or attenuated (Raichle, 2015).

To define individual differences in the connectivity pathway among scene-selective regions between good and poor imagers we used the VVIQ score as a covariate in PEB model analysis. We found that the influence of the signal from the left OPA to the ipsilateral PPA is positively related to the VVIQ score. The Vividness of Visual Imagery questionnaire includes items strictly related to the shape and the colour of an image that came into the mind’s eye. For instance: “think of a country scene which involves trees, mountains, and a lake. Consider the picture that comes before your mind’s eye: the contours of the landscape” and then “the colour and shape of the lake” (Marks, 1973). Likewise, it asked for scenes that the subject know well: “Think of the front of a shop which you often go to. Consider the picture that comes before your mind’s eye: The overall appearance of the shop from the opposite side of the road”(Marks, 1973).

It is well known that the analysis of category-related information takes place in the cortical regions of the ventral visual pathway (VVP) (Epstein and Kanwisher, 1998). In this part of the brain, the recognition of higher visual features takes place to distinguish, for example, a face from a place. Indeed, in the VVP are located both the Fusiform Face Area (FFA), responding preferentially to faces, and the Parahippocampal Place Area. Additionally, colour-selective regions, responding to coloured but not to black and white images, are interposed between FFA and PPA (Lafer-Sousa et al., 2016). It is plausible to hypothesize that the more the visual area OPA, that visually guides the low-level feature of navigation, is connected with the higher-order visual area PPA, the better the peculiar features of a scene are recognized. We could expect the same positive relation between the VVIQ score and the connectivity into other regions of the ventral visual pathway, such as FFA, for different category-specific stimuli (faces), but further studies are required.

Moreover, the ventral area PPA is known to respond more to scene categorization, distinguishing from an indoor to an outdoor place (e.g., a kitchen from a beach), but not to the change of view in a scene, unlike the more medial area RSC. Here, we found that the outgoing connectivity from OPA to RSC positively correlated with VVIQ in the left hemisphere, but the posterior probability didn’t exceed the threshold (pp = 0.85). The retrosplenial complex is the region that, more than the other scene-selective regions, is essential for building a “cognitive map” of the surrounding space, elaborating an egocentric view of the scene (Dilks et al., 2011; Persichetti and Dilks, 2016). Neuropsychological studies demonstrate that selective lesions to PPA or RSC cause different impairments in scene elaboration, suggesting that these regions have a complementary role in navigation. Precisely, lesions to PPA seem to be associated with topographical disorientation, in some cases anterograde and retrograde (Takahashi and Kawamura, 2002): patients are not able to recognize known and novel scenes and buildings. On the other hand, RSC is damaged, patients lose their way in familiar or novel environments with an inability to access the cognitive map of the place.

Furthermore, individual differences in cognitive style affect the type of information to be chosen, verbal or visual, for guiding navigation (Pazzaglia and De Beni, 2001) and good imagery abilities are crucial to this aim. In this study, we considered regions involved in scene recognition and we studied whether the imagery ability could be associated with the intrinsic connectivity among the selected regions. We found that the outgoing connectivity in the right hemisphere regions were not related to the VVIQ (apart from the connectivity from OPA to PPA that negatively correlates with VVIQ score). We suppose that the right hemisphere could have different connectivity with the more medial regions, such as the hippocampus, or frontal regions during imagery of places. Previously, Boccia et al. (2017b) using psychophysiological interaction (PPI) analysis, found that right PPA showed higher connectivity with right RSC during perception and higher connectivity with the right HC during imagery of familiar landmarks.

We are aware that imagery involves a wide set of regions so that more studies are required to investigate other aspects of imagery abilities. Moreover, further research is needed to investigate the effect of scene-selective regions on the hippocampus, in task-based as well as resting-state designs. For this reason, there are some limitations in our study which can be addressed in future. First, we used volume-based ROIs centred on the local maximum of the region that was nearest the maxima in the group mean map as criteria used to choose scene-selective regions. A surface-based approach would better define regions that are functionally defined. Moreover, we analysed data from subjects during resting-state fMRI so that our results show the intrinsic interactions among scene-selective regions. We are aware that results from task-based effective connectivity among scene-selective regions might vary from results we found in this study. In our case, dynamic causal modelling for resting state is based on cross-spectral density and has no experimental design regressors, while results on task-based DCM strictly depend on the type of stimulus used and the task performed.

In conclusion, our results provide new information about the causal interaction among brain regions involved in the perception and the imagery of scene/place and demonstrate that intrinsic fluctuations of signals among scene-selective regions reflect individual imagery abilities. Generally, understanding the complexity of the reciprocal influences among human brain regions is one of the most challenging goals for future neuroscientists. In our study, we decided to focus on regions belonging to the navigational system providing new insight in understanding how these regions communicate with each other. At the same time, since navigation is also dependent on imagery ability, we decided to test the relation of effective connectivity results with a behavioural measure. Our results confirm that the intrinsic connectivity among regions could be a key to explain the individual differences in navigational and possibly in other abilities.

Conflicts of interest

We have no conflict of interest to declare.

Availability of data and material

Data are available upon request to the corresponding author in compliance with the institutional ethics approval.

Code availability

Scripts used for running experiments and data analyses are included in SPM12.

CRediT authorship contribution statement

Maria Giulia Tullo: Conceptualization, Methodology, Formal analysis, Visualization, Writing- Original Draft. Hannes Almgren: Conceptualization, Methodology, Formal analysis, Software, Writing - Review & Editing. Frederick Van der Steen: Conceptualization, Methodology, Formal analysis, Software, Writing - Review & Editing. Valentina Sulpizio: Conceptualization, Methodology, Writing - Review & Editing. Daniele Marinazzo: Methodology, Supervision. Gaspare Galati: Methodology, Resources, Writing - Review & Editing, Supervision, Project administration.

Acknowledgments

The present study has been supported by funding from Sapienza University of Rome (PhD program in Behavioral Neuroscience) for MGT and by the Special Research Fund of Ghent University awarded to HA (grant No. BOF16/DOC/282; https://www.ugent.be/).

Auger SD, Maguire EA (2013) Assessing the mechanism of response in the retrosplenial cortex of good and poor navigators. Cortex 49:2904–2913. https://doi.org/10.1016/j.cortex.2013.08.002
Auger SD, Mullally SL, Maguire EA (2012) Retrosplenial Cortex Codes for Permanent Landmarks. PLoS ONE 7:e43620. https://doi.org/10.1371/journal.pone.0043620
Baldassano C, Beck DM, Fei-Fei L (2013) Differential connectivity within the Parahippocampal Place Area. NeuroImage 75:228–237. https://doi.org/10.1016/j.neuroimage.2013.02.073
Blazhenkova O, Kozhevnikov M (2010) Visual-object ability: A new dimension of non-verbal intelligence. Cognition 117:276–301. https://doi.org/10.1016/j.cognition.2010.08.021
Blazhenkova O, Kozhevnikov M (2009) The new object-spatial-verbal cognitive style model: Theory and measurement. Appl Cognit Psychol 23:638–663. https://doi.org/10.1002/acp.1473
Boccia M, Piccardi L, Palermo L, Nemmi F, Sulpizio V, Galati G, Guariglia C (2015) A penny for your thoughts! patterns of fMRI activity reveal the content and the spatial topography of visual mental images: Neural Correlates of Visual Mental Images. Hum Brain Mapp 36:945–958. https://doi.org/10.1002/hbm.22678
Boccia M, Sulpizio V, Nemmi F, Guariglia C, Galati G (2017a) Direct and indirect parieto-medial temporal pathways for spatial navigation in humans: evidence from resting-state functional connectivity. Brain Struct Funct 222:1945–1957. https://doi.org/10.1007/s00429-016-1318-6
Boccia M, Sulpizio V, Palermo L, Piccardi L, Guariglia C, Galati G (2017b) I can see where you would be: Patterns of fMRI activity reveal imagined landmarks. NeuroImage 144:174–182. https://doi.org/10.1016/j.neuroimage.2016.08.034
Bonner MF, Epstein RA (2017) Coding of navigational affordances in the human visual system. Proc Natl Acad Sci USA 114:4793–4798. https://doi.org/10.1073/pnas.1618228114
Burles F, Slone E, Iaria G (2017) Dorso-medial and ventro-lateral functional specialization of the human retrosplenial complex in spatial updating and orienting. Brain Struct Funct 222:1481–1493. https://doi.org/10.1007/s00429-016-1288-8
Committeri G, Galati G, Paradis A-L, Pizzamiglio L, Berthoz A, LeBihan D (2004) Reference Frames for Spatial Cognition: Different Brain Areas are Involved in Viewer-, Object-, and Landmark-Centered Judgments About Object Location. J Cogn Neurosci 16:1517–1535. https://doi.org/10.1162/0898929042568550
Dijkstra N, Zeidman P, Ondobaka S, van Gerven MAJ, Friston K (2017) Distinct Top-down and Bottom-up Brain Connectivity During Visual Perception and Imagery. Sci Rep 7:5677. https://doi.org/10.1038/s41598-017-05888-8
Dilks DD, Julian JB, Kubilius J, Spelke ES, Kanwisher N (2011) Mirror-Image Sensitivity and Invariance in Object and Scene Processing Pathways. J Neurosci 31:11305–11312. https://doi.org/10.1523/JNEUROSCI.1935-11.2011
Epstein R, Kanwisher N (1998) A cortical representation of the local visual environment. Nature 392:598–601. https://doi.org/10.1038/33402
Epstein RA, Baker CI (2019) Scene Perception in the Human Brain. Annu Rev Vis Sci 5:373–397. https://doi.org/10.1146/annurev-vision-091718-014809
Friston KJ, Harrison L, Penny W (2003) Dynamic causal modelling. NeuroImage 19:1273–1302. https://doi.org/10.1016/S1053-8119(03)00202-7
Friston KJ, Kahan J, Biswal B, Razi A (2014) A DCM for resting state fMRI. NeuroImage 94:396–407. https://doi.org/10.1016/j.neuroimage.2013.12.009
Friston KJ, Litvak V, Oswal A, Razi A, Stephan KE, van Wijk BCM, Ziegler G, Zeidman P (2016) Bayesian model reduction and empirical Bayes for group (DCM) studies. NeuroImage 128, 413–431. https://doi.org/10.1016/j.neuroimage.2015.11.015
Fulford J, Milton F, Salas D, Smith A, Simler A, Winlove C, Zeman A (2018) The neural correlates of visual imagery vividness – An fMRI study and literature review. Cortex 105:26–40. https://doi.org/10.1016/j.cortex.2017.09.014
Galati G, Pelle G, Berthoz A, Committeri G (2010) Multiple reference frames used by the human brain for spatial perception and memory. Exp Brain Res 206:109–120. https://doi.org/10.1007/s00221-010-2168-8
Ganis G, Thompson WL, Kosslyn SM (2004) Brain areas underlying visual mental imagery and visual perception: an fMRI study. Cogn Brain Res 20:226–241. https://doi.org/10.1016/j.cogbrainres.2004.02.012
Ino T, Inoue Y, Kage M, Hirose S, Kimura T, Fukuyama H (2002) Mental navigation in humans is processed in the anterior bank of the parieto-occipital sulcus. Neurosci Lett 322:182–186. https://doi.org/10.1016/S0304-3940(02)00019-8
Julian JB, Ryan J, Hamilton RH, Epstein RA (2016) The Occipital Place Area Is Causally Involved in Representing Environmental Boundaries during Navigation. Curr Biol 26:1104–1109. https://doi.org/10.1016/j.cub.2016.02.066
Kamps FS, Lall V, Dilks DD (2016) The occipital place area represents first-person perspective motion information through scenes. Cortex 83:17–26. https://doi.org/10.1016/j.cortex.2016.06.022
Keogh R, Pearson J (2014) The sensory strength of voluntary visual imagery predicts visual working memory capacity. Journal of Vision 14:7–7. https://doi.org/10.1167/14.12.7
Keogh R, Pearson J (2011) Mental Imagery and Visual Working Memory. PLoS ONE 6:e29221. https://doi.org/10.1371/journal.pone.0029221
Kosslyn SM, Ganis G, Thompson WL (2001) Neural foundations of imagery. Nat Rev Neurosci 2:635–642. https://doi.org/10.1038/35090055
Kraemer DJM, Schinazi VR, Cawkwell PB, Tekriwal A, Epstein RA, Thompson-Schill SL (2017) Verbalizing, visualizing, and navigating: The effect of strategies on encoding a large-scale virtual environment. Journal of Experimental Psychology: Learning Memory Cognition 43:611–621. https://doi.org/10.1037/xlm0000314
Kwong KK, Belliveau JW, Chesler DA, Goldberg IE, Weisskoff RM, Poncelet BP, Kennedy DN, Hoppel BE, Cohen MS, Turner R (1992) Dynamic magnetic resonance imaging of human brain activity during primary sensory stimulation. Proceedings of the National Academy of Sciences 89, 5675–5679. https://doi.org/10.1073/pnas.89.12.5675
Lafer-Sousa R, Conway BR, Kanwisher NG (2016) Color-Biased Regions of the Ventral Visual Pathway Lie between Face- and Place-Selective Regions in Humans, as in Macaques. J Neurosci 36:1682–1697. https://doi.org/10.1523/JNEUROSCI.3164-15.2016
Maguire E (2001) The retrosplenial contribution to human navigation: A review of lesion and neuroimaging findings. Scand J Psychol 42:225–238. https://doi.org/10.1111/1467-9450.00233
Marchette SA, Yerramsetti A, Burns TJ, Shelton AL (2011) Spatial memory in the real world: long-term representations of everyday environments. Mem Cogn 39:1401–1408. https://doi.org/10.3758/s13421-011-0108-x
Margulies DS, Vincent JL, Kelly C, Lohmann G, Uddin LQ, Biswal BB, Villringer A, Castellanos FX, Milham MP, Petrides M (2009) Precuneus shares intrinsic functional architecture in humans and monkeys. Proceedings of the National Academy of Sciences 106, 20069–20074. https://doi.org/10.1073/pnas.0905314106
Marks DF, 1973. VISUAL IMAGERY DIFFERENCES IN, THE RECALL OF PICTURES. Br J Psychol 64, 17–24. https://doi.org/10.1111/j.2044-8295.1973.tb01322.x
Mazziotta JC, Toga AW, Evans A, Fox P, Lancaster J (1995) A Probabilistic Atlas of the Human Brain: Theory and Rationale for Its Development. NeuroImage 2, 89–101. https://doi.org/10.1006/nimg.1995.1012
Mechelli A (2004) Where Bottom-up Meets Top-down: Neuronal Interactions during Perception and Imagery. Cereb Cortex 14:1256–1265. https://doi.org/10.1093/cercor/bhh087
Nasr S, Devaney KJ, Tootell RBH (2013) Spatial encoding and underlying circuitry in scene-selective cortex. NeuroImage 83:892–900. https://doi.org/10.1016/j.neuroimage.2013.07.030
Nasr S, Liu N, Devaney KJ, Yue X, Rajimehr R, Ungerleider LG, Tootell RBH (2011) Scene-Selective Cortical Regions in Human and Nonhuman Primates. J Neurosci 31:13771–13785. https://doi.org/10.1523/JNEUROSCI.2792-11.2011
Nori R, Giusberti F (2006) Predicting cognitive styles from spatial abilities. Am J Psychol 119:67–86
Oldfield RC (1971) The assessment and analysis of handedness: The Edinburgh inventory. Neuropsychologia 9:97–113. https://doi.org/10.1016/0028-3932(71)90067-4
Pazzaglia F, De Beni R (2001) Strategies of processing spatial information in survey and landmark-centred individuals. European Journal of Cognitive Psychology 13:493–508. https://doi.org/10.1080/09541440125778
Pazzaglia F, Moè A (2013) Cognitive styles and mental rotation ability in map learning. Cogn Process 14:391–399. https://doi.org/10.1007/s10339-013-0572-2
Pearson J (2019) The human imagination: the cognitive neuroscience of visual mental imagery. Nat Rev Neurosci 20:624–634. https://doi.org/10.1038/s41583-019-0202-9
Pearson J, Westbrook F (2015) Phantom perception: voluntary and involuntary nonretinal vision. Trends in Cognitive Sciences 19:278–284. https://doi.org/10.1016/j.tics.2015.03.004
Persichetti AS, Dilks DD (2016) Perceived egocentric distance sensitivity and invariance across scene-selective cortex. Cortex 77:155–163. https://doi.org/10.1016/j.cortex.2016.02.006
Piccardi L, Bocchi A, Palmiero M, Verde P, Nori R (2017) Mental imagery skills predict the ability in performing environmental directional judgements. Exp Brain Res 235:2225–2233. https://doi.org/10.1007/s00221-017-4966-8
Raichle ME (2015) The Brain’s Default Mode Network. Annu Rev Neurosci 38:433–447. https://doi.org/10.1146/annurev-neuro-071013-014030
Ranganath C, D’Esposito M (2005) Directing the mind’s eye: prefrontal, inferior and medial temporal mechanisms for visual working memory. Curr Opin Neurobiol 15:175–182. https://doi.org/10.1016/j.conb.2005.03.017
Silson EH, Chan AW-Y, Reynolds RC, Kravitz DJ, Baker CI (2015) A Retinotopic Basis for the Division of High-Level Scene Processing between Lateral and Ventral Human Occipitotemporal Cortex. J Neurosci 35:11921–11935. https://doi.org/10.1523/JNEUROSCI.0137-15.2015
Sulpizio V, Boccia M, Guariglia C, Galati G (2016) Functional connectivity between posterior hippocampus and retrosplenial complex predicts individual differences in navigational ability: Hippocampal Functional Connectivity Predicts Navigational Abilities. Hippocampus 26:841–847. https://doi.org/10.1002/hipo.22592
Takahashi N, Kawamura M (2002) Pure Topographical Disorientation —The Anatomical Basis of Landmark Agnosia. Cortex 38:717–725. https://doi.org/10.1016/S0010-9452(08)70039-X
Tullo MG, Sulpizio V, Boccia M, Guariglia C, Galati G (2018) Individual differences in mental imagery are predicted by the intrinsic functional architecture of scene-selective regions., in: Proceedings of Congress of the SEPEX, SEPNECA and AIP Experimental. Madrid
Wolbers T (2005) Dissociable Retrosplenial and Hippocampal Contributions to Successful Formation of Survey Representations. J Neurosci 25:3333–3340. https://doi.org/10.1523/JNEUROSCI.4705-04.2005
Xia M, Wang J, He Y (2013) BrainNet Viewer: a network visualization tool for human brain connectomics. PLoS One 8:e68910. https://doi.org/10.1371/journal.pone.0068910
Zeidman P, Jafarian A, Corbin N, Seghier ML, Razi A, Price CJ, Friston KJ (2019) A guide to group effective connectivity analysis, part 1: First level analysis with DCM for fMRI. NeuroImage 200:174–190. https://doi.org/10.1016/j.neuroimage.2019.06.031

Download PDF

Journal Publication

published 21 Mar, 2022

Read the published version in Brain Structure and Function →

Reviews received at journal
14 Sep, 2021
Editor invited by journal
10 Sep, 2021
Editor assigned by journal
09 Sep, 2021
First submitted to journal
07 Sep, 2021

You are reading this latest preprint version

Individual Differences in Mental Imagery Modulate Effective Connectivity of Scene-Selective Regions During Resting State

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Method

Participants

Experimental paradigm

Vividness of Visual Imagery Questionnaire

Image and acquisition processing

Region of Interest definition

Dynamic Causal Modelling (DCM) Analysis

Results

Group average connectivity strengths

Connectivity strengths and the relationship with VVIQ score

Discussion

Conclusions

Declarations

References

Status:

Journal Publication

Version 1