Marcus Robert Watson

Marcus Robert Watson

Auto-generated profile, based on publicly available information

Marcus Robert Watson

York University

Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
Verified
Journal Article
Open access
BVBenjamin Voloh
Benjamin Voloh
MWMarcus R Watson
Marcus R Watson
SKSeth Konig
Seth Konig
TWThilo Womelsdorf
Thilo Womelsdorf
Saccade detection is a critical step in the analysis of gaze data. A common method for saccade detection is to use a simple threshold for velocity or acceleration values, which can be estimated from the data using the mean and standard deviation. However, this method has the downside of being influenced by the very signal it is trying to detect, the outlying velocities or accelerations that occur during saccades. We propose instead to use the median absolute deviation (MAD), a robust estimator of dispersion that is not influenced by outliers. We modify an algorithm proposed by Nyström and colleagues, and quantify saccade detection performance in both simulated and human data. Our modified algorithm shows a significant and marked improvement in saccade detection - showing both more true positives and less false negatives – especially under higher noise levels. We conclude that robust estimators can be widely adopted in other common, automatic gaze classification algorithms due to their ease of implementation.

1 October 2019 in Journal of Neuroscience Methods
Verified
Journal Article
Closed access
MWMarcus R Watson
Marcus R Watson
BVBenjamin Voloh
Benjamin Voloh
CTChristopher Thomas
Christopher Thomas
AHAsif Hasan
Asif Hasan
TWThilo Womelsdorf
Thilo Womelsdorf

1 February 2017 in Consciousness and Cognition
Verified
Journal Article
Closed access
MWMarcus R. Watson
Marcus R. Watson
JCJan Chromý
Jan Chromý
LCLyle Crawford
Lyle Crawford
DEDavid M. Eagleman
David M. Eagleman
JEJames Enns
James Enns
KAKathleen A. Akins
Kathleen A. Akins

7 August 2018 in Behavior Research Methods
Verified
Journal Article
Open access
PDF file
MWMarcus R. Watson
Marcus R. Watson
BVBenjamin Voloh
Benjamin Voloh
MNMilad Naghizadeh
Milad Naghizadeh
TWThilo Womelsdorf
Thilo Womelsdorf
Many studies of vision and cognition require novel three-dimensional object sets defined by a parametric feature space. Creating such sets and verifying that they are suitable for a given task,...

13 December 2019 in PsyArXiv
Verified
Preprint
Open access
BVBenjamin Voloh
Benjamin Voloh
MWMarcus Watson
Marcus Watson
SKSeth König
Seth König
TWThilo Womelsdorf
Thilo Womelsdorf
Saccade detection is a critical step in the analysis of gaze data. A common method for saccade detection is to use a simple threshold for velocity or acceleration values, which is typically estimated from the data using the mean and standard deviation. However, this method has the downside of being influenced by the very signal it is trying to detect, the outlying velocities or accelerations that occur during saccades. We propose instead to use the median absolute deviation (MAD), a robust estimator of the standard deviation that is not influenced by outliers. We modify an algorithm proposed by Nyström and colleagues, and quantify saccade detection performance in both simulated and human data. Our modified algorithm shows a significant and marked improvement in saccade detection, showing both more true positives and less false negatives. We conclude that robust estimators can be widely adopted in other common, automatic gaze classification algorithms due to their ease of implementation.

5 October 2018 in bioRxiv
Verified
Preprint
Open access
PDF file
MWMarcus Robert Watson
Marcus Robert Watson
VBVoloh Benjamin
Voloh Benjamin
TCThomas Christopher
Thomas Christopher
HAHasan Asif
Hasan Asif
WTWomelsdorf Thilo
Womelsdorf Thilo
1 1.1 Background: There is a growing interest in complex, active, and immersive behavioral neuroscience tasks. However, the development and control of such tasks present unique challenges. 1.2 New Method: The Unified Suite for Experiments ( USE ) is an integrated set of hardware and software tools for the design and control of behavioral neuroscience experiments. The software, developed using the Unity video game engine, supports both active tasks in immersive 3D environments and static 2D tasks used in more traditional visual experiments. The custom USE SyncBox hardware, based around an Arduino Mega2560 board, integrates and synchronizes multiple data streams from different pieces of experimental hardware. The suite addresses three key issues with developing cognitive neuroscience experiments in Unity: tight experimental control, accurate sub-ms timing, and accurate gaze target identification. 1.3 Results: USE is a flexible framework to realize experiments, enabling (i) nested control over complex tasks, (ii) flexible use of 3D or 2D scenes and objects, (iii) touchscreen-, button-, joystick- and gaze-based interaction, and (v) complete offline reconstruction of experiments for post-processing and temporal alignment of data streams. 1.4 Comparison with Existing Methods: Most existing experiment-creation tools are not designed to support the development of video-game-like tasks. Those that do use older or less popular video game engines as their base, and are not as feature-rich or enable as precise control over timing as USE. 1.5 Conclusions: USE provides an integrated, open source framework for a wide variety of active behavioral neuroscience experiments using human and nonhuman participants, and artificially-intelligent agents. 2 Glossary: Active task : Experimental tasks which involve some combination of realistic, usually moving, stimuli, continuous opportunities for action, ecologically valid tasks, complex behaviours, etc. Here, they are contrasted with static tasks (see below) Arduino : A multi-purpose generic micro-processor, here used to control inter-device communication and time synchronization. Raycast : A game-engine method that sends a vector between two points in a virtual three-dimensional environment, and returns the first object in that environment it hits. Often used to determine if a character in a game can see or shoot another character. State Machine (also Finite State Machine) : A way of conceptualizing and implementing control in software, such that at any one moment the software is in one, and only one, state. In hierarchical state machines , as used in the present software suite, these can be organized into different levels, such that each level can only be in one state, but a state can pass control to a lower level. Static task : Experimental tasks like those traditionally used in the cognitive neurosciences. Simple, usually stationary, stimuli, limited opportunities for action, simple behaviours, etc. Here, they are contrasted with active tasks (see above) . Unity : One of the most popular video game engines. Freely available. Video game engine : A software development kit designed to handle many of the common issues involved in creating video games, such as interfacing with controllers, simulating physical collisions and lighting, etc.