Neural mechanisms of long-range spatial vision: an investigation of perceptive, integrative and association fields across the lifespan

Lead Research Organisation: University of Aberdeen
Department Name: Psychology

Abstract

Incoming light is processed by retinal receptors. Light reflected from objects that are nearby each other will fall onto adjacent retinal locations, and will be fed to correspondingly located neurons in the visual cortex. In the visual cortex, units of different complexity will process this information further, ultimately giving rise to our everyday visual experience. The area of space on the retina to which a visual cortex unit responds is called a visual field. The concept of visual fields has played a crucial role in the study of visual perception since the pioneering studies of Hubel and Wiesel in the 1960s. Visual fields determine which information will be bound together and which will be individuated. Away from the centre of the retina, visual fields increase in size, but this size is flexible and, somewhat surprisingly, depends on the strength of the received signals (colour or brightness) as well as the presence or absence of nearby objects, which activate neighbouring neural units. There are also visual fields of different complexity: while neurons in the primary visual cortex simply integrate the amount of brightness or colour contrast that fall within their scope, neurons beyond it integrate or individuate more complex attributes of objects, e.g. orientation, to identify the object's shape. Therefore, visual fields determine information processing across space, especially processing of relatively long-range spatial information, which will be at some distance from each other once we move away from the centre of the retina.

The aim of this project is to build a unitary framework that would encompass neural processing within visual fields of different complexity across the lifespan. Despite years of research, such a unitary framework is still lacking: while we know a lot about "low-level" processing of colour and luminance supported by neurons in the primary visual cortex, the intermediate stages of visual processing are more difficult to probe and distinct aspects of such "mid-level" vision are often studied separately. The uniqueness of our approach is that we will probe visual fields of different complexity, covering both "low" and "mid-level" vision, using the same paradigm. This paradigm is simple yet powerful: it relies on the same stimulus, consisting of multiple oriented elements (black and white gratings known as Gabor patches) that can form different letters, to probe visual fields of different complexity by simply changing the task that the participants are performing: a) detecting the presence or absence of a target element, b) judging its properties, or c) judging the identity of the whole letter. We will establish how visual fields process stimuli defined by brightness, colour, or both, thus integrating all types of contrast visible to the human eye. Finally, we will conduct our experiments on a large sample that consists of participants that are between 20 and 80 years of age. Our experiments will combine behavioural and neuroscientific techniques, using new, state-of-the-art electroencephalographic (EEG) techniques to ascertain which neural mechanisms underlie age-related changes in visual processing. We will determine the degree to which age-related deficits are driven by an increase in neural noise, and relate it to more basic changes in contrast sensitivity. Such analyses of age-related differences will offer an exciting window into the neural mechanisms that sustain information processing in human vision. In this way, this project will not only provide a basis for theoretical developments in visual perception, but will also considerably affect our understanding of the aging of sensory mechanisms. Such information can be used to improve the quality of life for the elderly.

Technical Summary

Human vision contains several kinds of visual fields, which organise different types of information. Receptive fields take in basic signals on brightness and colour from the environment, while integrative and association fields constrain more advanced, feature-based processes crucial for perceptual organisation of visual scenes. Similarities and differences between neural processes that occur within different kinds of visual fields have posed an important research question for many years, but despite that we still lack a robust unitary framework that assesses neural processing within all three types of visual fields. Using an established paradigm developed by the investigators, which can probe different visual fields with the same stimulus, this project intends to combine behavioural and neuroscientific techniques to systematically examine processes occurring within receptive, integrative and association fields. We will determine their common processing constraints using stimuli defined by both luminance and colour contrast. We will also trace age-related changes in their function using a large cohort of participants between 20 and 80 years old. The participants will be without neurological or visual deficits, allowing for a thorough insight into healthy aging. We will use powerful, new single-trial EEG methods that are able to quantify both the strength and the variability of neural activity. In this way, we will be able to assess the neural mechanisms that underpin processing of visual information within visual fields of different complexity. By contrasting the strength and variability of neural activity, we will determine the degree to which the neural inhibition model or the information degradation hypothesis can account for the age-related changes in the visual system.

Planned Impact

We have identified the following beneficiaries:
Academic community: Primarily of interest to visual and cognitive neuroscientists, the outputs of this project will also appeal more broadly to the fields of psychology, computer science and optometry. Academic work will be presented at a variety of specialist and general neuroscience and vision science conferences, to ensure impact in all these fields. We will explore potential avenues for ophthalmic impact, e.g. in the early diagnosis of visual impairment, as our large dataset will provide normative data on healthy aging of visual mechanisms. Some of the proposed work will complement and feed into an active interdisciplinary project on spatio-chromatic appearance modelling (Wuerger & Martinovic, funded by EPSRC), whose aim is to specify models that are consistent and constrained by biological vision including chromatic processing at different spatial scales. We will further engage with academics from complementary disciplines as part of the workshops on real and virtual environment design, which we intend to organise at the end of the project. There will also be a positive impact on the UK research skill-base as the PDRA and the graduate RA will both obtain a high-level, multi-method research training and interdisciplinary experience. UK is a leader in fundamental research in the fields of cognitive and visual neurosciences and this hard-earned reputation will be reinforced by the projected academic impact of the proposed work.

The public: We will engage the public through science events and through the project's website, which will be regularly updated by the PI and PDRA, populated with Youtube videos and other accessible content. Public engagement activities will be facilitated by the University of Aberdeen's Public Engagement with Research unit and Press and Events Teams. The PI has considerable experience in public engagement and is a member of the STEM network. PDRA will be trained in public engagement by the PI and encouraged to join the STEM network.
Design industries: Even though we propose a fundamental neuroscience project, there are potential applications that may be of high value to the UK at large. In familiar environments humans often reach for objects without looking at them directly, e.g. when grasping a bar of soap in a shower from a small holder cluttered with a facial wash, shampoo, and conditioner. However, failures to perceive due to insufficient object individuation can lead to failures in correct action; bending to reach for a dropped item may lead to falls. Failures to detect information on websites due to reduced visibility caused by clutter are also higher in older adults and can lead to frustration and disengagement with technology. We intend to actively seek applications for our work in the domains of real and artificial environmental design.

Through the North of Scotland Knowledge Transfer Partnership (KTP) centre, we have obtained contact with local experts, including those involved in designing bathrooms in residences for the elderly. The KTP centre will further help to invite relevant stakeholders to our impact workshops, which we will hold to ensure the application of our results. To get a better impression of the challenges posed by visual changes occurring in the healthy ageing population (e.g. reduced ability to see in low lighting, glare sensitivity) we will organise yearly focus groups, recruiting older adults from the School of Psychology participant panel. In essence, these focus groups are also another, more detailed way of engaging with the public. We will also contact organisations representing partially sighted individuals to canvas their views and needs on improving ease-of-access both in terms of interior design and display technologies. Ultimately, the outcome of our work could lead to improvements to "Good Practice in Design for Dementia and Sight Loss", which is a set of guidelines for the use of colour in design.
 
Description We developed a novel stimulus paradigm that can test processes occurring within receptive, integrative and association fields and collected a rich dataset from an age-diverse sample of healthy, normally sighted participants. Receptive fields subserve basic detection of information - e.g. is there a line on the screen? Individuating properties of elements occurs in integration fields - e.g., what is the orientation of the line? Meanwhile, binding of neighbouring elements into figures occurs within association fields - do the lines form a shape? Currently, we have good models of low-level, basic vision, but models of individuating and binding (so called mid-level vision, or perceptual organisation) are less well specified. Similarly, a model that fully connects low-level and mid-level vision is lacking. Our studies represent a significant step towards this model. We found that computations performed by each type of visual field operate over different distances - with receptive fields being the smallest and association fields being the largest. This is in line with previous research. In addition to that, we confirmed that sizes of visual fields co-varied across observers, but not across retinal locations. In other words, those people who have larger receptive fields will also have larger integrative fields, but the size of these fields is relatively stable and independent across the retina. Finally, we discovered co-dependencies between different field types across different retinal locations in terms of how they depend on contrast. Contrast dependency of detection and individuation was positively correlated across the retina. On the other hand, individuation and binding showed opposite patterns of contrast dependency - binding needed less contrast than individuation closer to central vision, but more contrast than individuation in peripheral vision. This means that a basic, low-level bottleneck does indeed lie at the core of mid-level vision, constraining information processing across different units - receptive, integration and association fields. Therefore, to fully understand mid-level vision, contrast needs to be incorporated into models and parametrically modulated in future experimental and modelling work. In a series of separate experiments, we confirmed the importance of this contrast bottleneck for individuation of element orientation. At low contrasts that are only just sufficient to discriminate orientation of single elements, addition of neighbouring elements reduces performance in a more general way which is independent of the distance of neighbours to the target. However, when contrasts are reliably above orientation discrimination threshold, a graded interference from neighbouring elements emerges, with closer elements interfering more than further elements. Understanding the interface between low-level, basic vision and mid-level vision under conditions of reduced contrast will not only allow for more refined models of vision but also lead to better understanding of age-related changes in the visual system, or deficits in low-vision patients.
Exploitation Route Models of visual perception that incorporate both low-level and mid-level vision stages will benefit greatly from our findings. As all our data will be publicly available upon acceptance of our manuscripts, this will enable further interrogation of the datasets as well as modelling work.
Sectors Digital/Communication/Information Technologies (including Software),Healthcare

 
Description Investigating the microstructure of human visual fields and generating low-vision applications
Amount £6,860 (GBP)
Funding ID Investigating the microstructure of human visual fields and generating low-vision applications 
Organisation University of Aberdeen 
Sector Academic/University
Country United Kingdom
Start 06/2019 
End 04/2021
 
Title Dataset from a large-scale experiment on low and mid-level vision 
Description Data from an age-diverse sample of 40 participants, with luminance contrast sensitivities and performance on masking, crowding and grouping tasks utilising a common stimulus framework. 
Type Of Material Database/Collection of data 
Year Produced 2022 
Provided To Others? Yes  
Impact The dataset will enable modelling of the contrast bottlenecks shared between low and mid-level vision by others, furthering progress into our understanding of these important stages of visual processing. 
URL https://osf.io/qd2bp/
 
Title Dataset from two experiments on visual crowding with low contrast stimuli (Lee, Reuther, Chakravarthi & Martinovic, 2021) 
Description This is the dataset that forms the basis of our pre-print "Emergence of crowding: The role of contrast and orientation salience". The dataset consists of two experiments. The first experiment evaluates crowding using S-cone islating and low-contrast luminance isolating stimuli with 8 experienced participants. The second experiment has a larger sample of 24 participants and examines crowding with low-contrast luminance isolating stimuli. The pre-print describes in detail the methods and results obtained from the experiment, and a 'readme' file in the data repository denotes the structure of the data. 
Type Of Material Database/Collection of data 
Year Produced 2021 
Provided To Others? Yes  
Impact The dataset allows for re-analysis of our findings or for meta-analytic work that intends to compile multiple datasets and perform additional tests that go beyond what is reported in our pre-print. 
URL https://osf.io/5a8ec/
 
Description Collaboration with Prof Daniel Coates on microstructure of human visual fields 
Organisation University of Houston
Department College of Optometry
Country United States 
Sector Academic/University 
PI Contribution We will provide EEG training to Prof Coates team during his visit to Aberdeen, postdoctoral researcher Dr Reuther will also provide her expertise during the visit to Houston to conduct AOSL experiments.
Collaborator Contribution Prof Coates will host Dr Reuther in Houston for a collaborative AOSL research project, which will be conducted on his adaptive optics system adjusted for testing S-cone function. He will also provide expertise on AMD.
Impact We obtained follow-up funding: BBSRC international partnering award for a project entitled "Investigating the microstructure of human visual fields and generating low-vision applications".
Start Year 2019
 
Description "Speed, brightness and noise" - digital talk at the Aberdeen Art Gallery 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Public/other audiences
Results and Impact "Speed, brightness and noise" discusses the inherently noisy nature of perception, and uses several demonstrations to show how this noisyness affects how we see and what we see. The talk connects these concepts to the work of Marcel Duchamp and was intended to resonate with conceptual artworks that were meant to be exhibited in the gallery in April 2020. Due to the Covid-19 lockdown, the talk was recorded and made available digitally, rather than being given in the gallery setting. 120 people have viewed the talk on Youtube and i received positive feedback from members of the public who commented that it had sparked an interest in vision science related topics.
Year(s) Of Engagement Activity 2020
URL https://www.youtube.com/watch?v=ahdvgApqiwM
 
Description Dundee Science Festival 2021 - Visual Illusions 
Form Of Engagement Activity Engagement focused website, blog or social media channel
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Public/other audiences
Results and Impact I generated online content on visual perception for the Dundee Science Festival, held in November 2021, including interactive demos that allowed participants to explore visual illusions in a scientific way, manipulating key parameters of the image to see how they influence what is seen. The activity enabled the audience to have a hands-on engagement with science from the safety of their own homes in a period when this would not have been possible in person due to the existing covid-19 regulations.
Year(s) Of Engagement Activity 2021
URL https://www.dundeesciencefestival.org/2021-11-12-visual-illusions-the-university-of-edinburgh
 
Description Interview for the Posifest project 
Form Of Engagement Activity Engagement focused website, blog or social media channel
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Public/other audiences
Results and Impact I recorded a Youtube interview with Shalhavit-Simcha Cohen, multimedia educator and organiser of Posifest - a festival of positive psychology. We discussed how thinking about the 'shortcuts' taken by our visual perception system to construct our everyday experiences can be a healthy paralel to encourage positive thinking about mental health and mechanisms that sustain it. This will build towards the bigger aim of Cohen's research, which aims to improve mental health of young adults by re-package academic research into accessible and social-media friendly formats.
Year(s) Of Engagement Activity 2021
URL https://www.youtube.com/watch?v=dQt7VVSVifc&t=28s
 
Description Participant panel research visit day - demonstrations of perceptual phenomena and discussions of potential impact 
Form Of Engagement Activity Participation in an open day or visit at my research institution
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Public/other audiences
Results and Impact Around 60 members of the older adult participant panel of the School of Psychology, University of Aberdeen, attended an annual 2-hour public engagement lunchtime event. Researchers discuss the outcomes of studies that involved panel volunteers or present various engaging and thought-provoking psychology-related demonstrations. While my current PhD student Ms Rozman (BBSRC Eastbio) and my postdoc Dr Reuther (BBSRC) discussed their own research with panel members and demonstrated various visual phenomena which we have prepared for such events (e.g. the Beuchet chair, the hollow face illusion, etc.), I discussed the EPSRC project on visual displays with interested panel members and handed out questionnaires for those who wanted to take part in the qualitative part of the project. Panel members were enthused about their views being given consideration by our project and I handed out more than 30 questionnaires and had conversations with more than 50 older adults. Due to the school holidays, quite a few of them brought their grandchildren to the event, who were enthralled by some of the visual demonstrations, in particular the opportunity to take family photos with the Beuchet chair illusion. This should engender enthusiasm about visual science in both the youngest and the older generation.
Year(s) Of Engagement Activity 2020
 
Description Visual Ageing Workshop 2021 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Postgraduate students
Results and Impact I organised a half-day workshop on visual ageing, bringing together researchers from different disciplines (psychology, architecture, user experience, display engineering) with an interest in this topic. The workshop began with talks from experts in the field (Prof Leonards, University of Bristol, Dr Allen, University of Nottingham, Dr Akhavan, Faurecia IRYSTec, Prof Platt, Harvard Extension School) which was followed by a general discussion and a networking session. Amongst the 30 attendees, the majority were postgraduates and postdoctoral researchers, mainly from the UK but also including some international participants. The attendees reported that the event benefited both their own research and their future work due to the increased awareness of interdisciplinarity, with some forming potential collaborative links with researchers from other disciplines during the event.
Year(s) Of Engagement Activity 2021
URL https://2021visualageing.wordpress.com/