Monkeys and apes, humans included, are naturally capable of complex coordinated tasks that befuddle state-of-the-art robots. A crucial component of such ability in humans and non-human priamtes is eye-hand coordination, which, as part of our common evolutionary heritage, is likely to have similar mechanisms and played a pivotal role in our survival. Although the importance of eye-hand coordination in activites like arranging lego blocks or reaching for a glass of water may not be obvious to many of us, it is a matter of life and death each time a monkey swings from branch to branch.
We usually make hand movements independently of eye movements. Yet when the task demands accuracy, the two can work together. “I think eye-hand coordination is a useful model to understand how the brain couples and decouples [its] modules,” says Prof. Aditya Murthy of the Centre for Neuroscience, Indian Institute of Science, Bangalore. “The general idea is to know how the brain generates this flexibility.” Using computer simulations and behavioural tasks performed by human subjects, Atul Gopal and Prof. Murthy have shown that a ‘common command’ architecture might be at work to execute coordinated tasks.
When a visual response is produced, two things happen: light from the object hits the retina and a signal is sent to the midbrain. Then another signal is transmitted through the motor neurons to make the eye move in response. Some reactions are quick, like this one. Yet for mentally demanding tasks, the typical response time is much longer. This ‘procrastination’ indicates the use of a longer/slower circuit during decision making. Thus, our reactions are a constant push-pull between a fast 'dumb' process and a slower 'intelligent' process. A whole body of work in this field has suggested mechanisms for coordination between eye and hand movements in isolation: neurons steadily increase their firing rate till they reach a threshold. This process produces the reaction time and the variability we observe.
There are three possible architectures to explain coordination based on the above mechanism. One is an interaction between of independent eye and hand systems that results in coordination simply because of the common stimulus. The second is that an interaction is established when coordination is required. The third, a computationally parsimonious model, has both motor commands coming from a common centre.
Gopal evaluated these architectures with the help of a well-known model in physics known as the diffusion model', which is generally used to understand decision-making. The key idea used here is that when reaction times are longer, there is more variation. This has been shown to hold individually. But in coordinated tasks, variation in hand reaction time is related to how quick the eye movement is. This is a peculiar variation, diagnostic of a ‘common command’ type of architecture.
“It is very simple, but nobody looked at this,” says Prof Murthy. This piece of evidence is also validated through the electromyograph (EMG) results of subjects while they performed certain tasks.
Furthering their investigations, the researchers have also shown the presence of an inhibitory system in the common command centres. A motor system must be able to produce as well as inhibit movements at will. Such a double-faced control mechanism is a hallmark of what is called “Executive” Control. When we want to learn a complicated task, we use different areas of the brain to have executive control; information received is restructured and an informed response is directed. The presence of a common inhibitory control that controls eye and hand movements simultaneously gives further credence to a dedicated motor system which the authors have claimed to be at work.
The experiments have also shed light on another fundamental question that has been long been a matter of debate, namely, whether our motor systems can present a distinct point of no return. Such a ballistic stage is a point below the threshold in the activity build-up beyond after which the response cannot be withdrawn. Scientists have long wondered whether and how such a ballistic stage might operate. During eye-hand movements there appears to be evidence of presence of a ballistic stage in hand movements. Taken together these studies reveal how the judicious application of simple behaviour measures such as reaction time can be used to develop and test computational models of brain function.
About the authors:
Prof Aditya Murthy is a faculty member at the Centre for Neuroscience at IISc Bangalore. Mr.AtulGopal is a graduate student at Prof Murthy’s lab. The authors can be contacted at firstname.lastname@example.org
The paper “Eye-hand coordination during a double-step task: evidence for a common stochastic accumulator” appeared in the Journal of Neurophysiology. It can be accessed here: doi:10.1152/jn.00276.2015.