Eirini Mavritsaki

The neural mechanisms of visual selection: the view from neuropsychology

In this review, we discuss how neuropsychological impairments in visual selection can inform us about how selection normally operates. Using neuroanatomical and behavioral evidence on the disorders of neglect, extinction, and simultanagnosia, we propose functional and anatomical links between different aspects of visual selection and distinct sites in the posterior parietal cortex (PPC). This includes linking: (i) bottom‐up attentional capture and the right temporo‐parietal junction (TPJ); (ii) top‐down segmentation of displays and the medial PPC; (iii) grouping, individuation and identification, and the inferior intra‐parietal sulcus (IPS) bilaterally; and (iv) the suppression of saliency and the left IPS. In addition, when neuropsychological studies are combined with fMRI, there is evidence that these regions of the PPC interact with striate and extra‐striate cortical areas, which respond to specific properties of stimuli. Selection should be viewed as an emergent property of a network of areas involving both ventral and dorsal cortex.

Decomposing the neural mechanisms of visual search through model-based analysis of fMRI: Top-down excitation, active ignoring and the use of saliency by the right TPJ

Despite being studied intensively over the past 30 years, the neural processes underlying visual search are not yet fully understood. In the current study we extend prior work using model-based analysis to decompose fMRI data. fMRI data on human search were assessed using activation functions predicted from the spiking Search over Time and Space model (sSoTS; Mavritsaki et al., 2006). Going beyond previous work, we show for the first time that activity in a central location map in the model, which computes the saliency of a target relative to distractors, correlated with the BOLD response in the right temporo-parietal junction (TPJ)--a key region implicated in clinical studies of unilateral neglect. This is consistent with the right TPJ responding to the relative saliency of visual stimuli. In addition, a re-analysis of search performance, with a larger participant set and a psychologically plausible response rule, showed distinct neural regions in parietal and occipital cortices linked to top-down excitation and the to active ignoring of distractors. The results indicate that excitatory and inhibitory circuits for visual selection can be separated, and that the right TPJ may be critical for responding to salient targets. The value of using a model-based approach is discussed.

Simulating posterior parietal damage in a biologically plausible framework: neuropsychological tests of the search over time and space model.

The search over time and space (sSoTS) model attempts to simulate both the spatial and the temporal aspects of human visual search using spiking level neurons, which incorporate some biologically plausible aspects of neuronal firing. The model contains pools of units that (a) code basic features of objects, presumed to reside in the ventral visual stream, and (b) respond in a feature-independent way to stimulation at their location, presumed to operate in the posterior parietal cortex. We examined the effects of selective lesioning neurons responding to one side of the location map. Unilateral damage introduced spatial biases into selection that affected conjunction more than single-feature search. In addition, there was an impaired ability to segment stimuli over time as well as space (e.g., in preview search). These results match previously reported data on patients with posterior parietal lesions. In addition we show that spatial biases in selection increase under conditions in which there is decreased activity from excitatory neurotransmitters, mimicking effects of reduced arousal. Further simulations explored the effects of time and of visual grouping on extinction, generating predictions that were then tested empirically. The model provides a framework for linking behavioural data from patients with neural-level determinants of visual attention.

Response linearity determined by recruitment strategy in detailed model of nictitating membrane control

Many models of eyeblink conditioning assume that there is a simple linear relationship between the firing patterns of neurons in the interpositus nucleus and the time course of the conditioned response (CR). However, the complexities of muscle behaviour and plant dynamics call this assumption into question. We investigated the issue by implementing the most detailed model available of the rabbit nictitating membrane response (Bartha and Thompson in Biol Cybern 68:135–143, 1992a and in Biol Cybern 68:145–154, 1992b), in which each motor unit of the retractor bulbi muscle is represented by a Hill-type model, driven by a non-linear activation mechanism designed to reproduce the isometric force measurements of Lennerstrand (J Physiol 236:43–55, 1974). Globe retraction and NM extension are modelled as linked second order systems. We derived versions of the model that used a consistent set of SI units, were based on a physically realisable version of calcium kinetics, and used simulated muscle cross-bridges to produce force. All versions showed similar non-linear responses to two basic control strategies. (1) Rate-coding with no recruitment gave a sigmoidal relation between control signal and amplitude of CR, reflecting the measured relation between isometric muscle force and stimulation frequency. (2) Recruitment of similar strength motor units with no rate coding gave a sublinear relation between control signal and amplitude of CR, reflecting the increase in muscle stiffness produced by recruitment. However, the system response could be linearised by either a suitable combination of rate-coding and recruitment, or by simple recruitment of motor units in order of (exponentially) increasing strength. These plausible control strategies, either alone or in combination, would in effect present the cerebellum with the simplified virtual plant that is assumed in many models of eyeblink conditioning. Future work is therefore needed to determine the extent to which motor neuron firing is in fact linearly related to the nictitating membrane response.

Using biologically plausible neural models to specify the functional and neural mechanisms of visual search.

We review research from our laboratory that attempts to pull apart the functional and neural mechanisms of visual search using converging, inter-disciplinary evidence from experimental studies with normal participants, neuropsychological studies with brain lesioned patients, functional brain imaging and computational modelling. The work suggests that search is determined by excitatory mechanisms that support the selection of target stimuli, and inhibitory mechanisms that suppress irrelevant distractors. These mechanisms operate through separable though overlapping neural circuits which can be functionally decomposed by imposing model-based analyses on brain imaging data. The chapter highlights the need for inter-disciplinary research for understanding complex cognitive processes at several levels.

Suppressive effects in visual search: A neurocomputational analysis of preview search

In the real world, visual information is selected over time as well as space, when we prioritise new stimuli for attention. Watson and Humphreys [Visual marking: prioritising selection for new objects by top-down attentional inhibition of old objects. Psychol. Rev. 104 (1997) 90-122] presented evidence that the prioritisation of new information in search tasks depends, at least in part, on the active ignoring of old items-a process they termed visual marking. In the present paper we present for the first time an explicit computational model of visual marking using a biologically plausible neural network. The model incorporates different synaptic components and a frequency adaptation mechanism, which acts to suppress the previously attended items. We show that, when coupled with a process of active inhibition to the old items, the pattern of preview search can be captured, as well as both efficient and inefficient search patterns in baseline conditions. The simulations point to the involvement of both active and passive inhibitory mechanisms in the preview effect in human search.

Model Based Analysis of fMRI-Data: Applying the sSoTS Framework to the Neural Basic of Preview Search

The current work aims to unveil the neural circuits underlying visual search over time and space by using a model-based analysis of behavioural and fMRI data. It has been suggested by Watson and Humphreys [31] that the prioritization of new stimuli presented in our visual field can be helped by the active ignoring of old items, a process they termed visual marking. Studies using fMRI link the marking process with activation in superior parietal areas and the precuneus [4,18,27,26]. Marking has been simulated previously using a neural-level account of search, the spiking Search over Time and Space (sSoTS) model, which incorporates inhibitory as well as excitatory mechanisms to guide visual selection. Here we used sSoTS to help decompose the fMRI signals found in a preview search procedure, when participants search for a new target whilst ignoring old distractors. The time course of activity linked to inhibitory and excitatory processes in the model was used as a regressor for the fMRI data. The results showed that different neural networks were correlated with top-down excitation and top-down inhibition in the model, enabling us to fractionate brain regions previously linked to visual marking. We discuss the contribution of model-based analysis for decomposing fMRI data.

Cross-cultural differences in perceptual learning

Cross-cultural studies have shown that independence in individualistic societies is associated with analytic systems of thoughts, whereas collectivistic societies which place greater emphasis on interdependence are generally predisposed to holistic thinking (Masuda & Nisbett, 2011; Bang, 2015). Further, cultural identity has been shown to effect picture perception and cognitive processes (Nisbett & Miyamoto, 2005). For example, Asians were shown to be more sensitive to contextual rather than focal information compared to Americans in a change-blindness task in which observers had to detect either a focal or contextual change within pairs of images (Masuda & Nisbett, 2006). Here, we build on previous work on perceptual learning (Mayhew, Li, & Kourtzi, 2012) and test the role of individualistic vs. collectivistic influences on learning ability. Eighty-three participants of different cultural backgrounds – consisting of Asian (collectivistic) and European (individualistic) students – were asked to discriminate between radial and concentric Glass patterns embedded in background noise. We employed the Singelis’ (1994) self-construal scale (SCS) to examine whether differences in task performance due to training were influenced by independent or interdependent cultural values. Visual perceptual learning was evident in both groups; that is, all participants improved in accuracy and reaction times through training. Importantly, Asian participants showed higher performance before and during training than European participants, suggesting an advantage in learning to extract global shapes embedded in cluttered backgrounds. This is consistent with the previously reported tendency of collectivists for global processing under perceptual uncertainty. Our findings provide evidence for the role of cross-cultural influences on visual processes that relate to our ability to improve in making perceptual judgements through training and experience. Keywords: visual perception, learning, self-construal,

Temporal binding and segmentation in visual search: A computational neuroscience analysis

Human visual search operates not only over space but also over time, as old items remain in the visual field and new items appear. Preview search (where one set of distractors appears before the onset of a second set) has been used as a paradigm to study search over time and space [Watson, D. G., & Humphreys, G. W. Visual marking: Prioritizing selection for new objects by top–down attentional inhibition of old objects. Psychological Review, 104, 90–122, 1997], with participants showing efficient search when old distractors can be ignored and new targets prioritized. The benefits of preview search are lost, however, if a temporal gap is introduced between a first presentation of the old items and the re-presentation of all the items in the search display [Kunar, M. A., Humphreys, G. W., & Smith, K. J. History matters: The preview benefit in search is not onset capture. Psychological Science, 14, 181–185, 2003a], consistent with the old items being bound by temporal onset to the new stimuli. This effect of temporal binding can be eliminated if the old items reappear briefly before the new items, indicating also a role for the memory of the old items. Here we simulate these effects of temporal coding in search using the spiking search over time and space model [Mavritsaki, E., Heinke, D., Allen, H., Deco, G., & Humphreys, G. W. Bridging the gap between physiology and behavior: Evidence from the sSoTS model of human visual attention. Psychological Review, 118, 3–41, 2011]. We show that a form of temporal binding by new onsets has to be introduced to the model to simulate the effects of a temporal gap, but that effects of the memory of the old item can stem from continued neural suppression across a temporal gap. We also show that the model can capture the effects of brain lesion on preview search under the different temporal conditions. The study provides a proof-of-principle analysis that neural suppression and temporal binding can be sufficient to account for human search over time and space.

Towards grasping the underlying neuronal processes in ADHD using a visual search task: a computational modelling approach.

Attention is essential for our everyday life. The role of attention is to identify the most relevant information in our environment to be processed. Deficits in attention processes have been linked to attention deficit hyperactivity disorder (ADHD). ADHD is a disorder with an onset in childhood and is characterised by inattentiveness, hyperactivity and impulsivity. In the present study we investigate the link between ADHD and visual attention. ADHD is linked with a reduction in arousal levels, which is related to norepinephrine and dopamine function. It is difficult to investigate the role of these two neurotransmitters in ADHD without using interventions. An attractive approach is to use computational modelling, as there is no need for interventions, if the appropriate model is used. In the present study, we extended a neural-level model, the spiking Search over Time and Space (sSoTS) model which was developed to simulate visual search [1]. The important characteristics of sSoTS that qualifies it as an appropriate tool is the incorporated top-down processes and the neuronal details of the system. We will present the outcomes of the first step of our study; using neurotransmitter changes to simulate ADHD behavioural results in a visual search experiment. Importantly the combination of simulated dopamine and norepinephrine function in the model allowed us to be able to stimulate the ADHD behavioural results. The next step in this work is to identify the mechanism(s) responsible for the deficit of attentional function in children with ADHD. REFERENCES [1] Mavritsaki, et al. (2011) Psychological Review, 118(1): p.3-41. Meeting abstract presented at VSS 2015.