Which apple is the heaviest?
Our brain can't wait to get the information via our sensors (eye, skin, muscle stretch-proprioception, ...) because they are too slow (processing time of at least 100ms). Early visual signals arrive in the visual cortex in 40ms. Additional processing by other visual areas refine the analysis of the visual scene but also increases the processing time. For instance, the unambiguous direction of a moving stimulus will take as much as 80ms (Pack and Born, 2001). In 80ms, a car driving at 120km/h will have moved by more than 25m!!!! Therefore, prediction represents one way to circumvent the low speed of our sensors.
The speed of a falling object increases with time as gravity will accelerate it. Therefore, prediction of object motion is only possible if gravity is taken into account. To do so, the brain uses internal representation of object motion in order to predict future target trajectory. There is a reasonable consensus about the existence of such internal representation of object motion (Zago et al. 2009) and about its use in order to guide smooth pursuit eye movements (Orban de Xivry et al. 2008 and many others).
The explanation for this phenomena relies on the ability of the brain to predict how and when your own hand will pick up the book but is unable to predict when the other person will do so. The same type of prediction is responsible for the fact that you are unable to tickle yourself (Blakemore et al. 2000) because your brain predicts the effect of your fingers on your waist and is able to cancel the sensation related to the tickling. Note that many species use the same kind of predictions for sensory cancellation (attenuation of the sensation linked to your own actions). For instance, crickets do not become mad even if they sing all day long because they are able to predict the noise that they're gonna produce and are able to cancel it (see Webb 2004 for many more examples). For the same reason, your own voice always sounds strange when you hear yourself on a video. Indeed, your brain filters out your own voice in order to be able to perceive other more relevant sounds when you are talking (and it allows you to listen to the other person).
The design of our brain results from years of evolution. In small vertebrates, from which we evolved, there were potentially only a few neurons that needed to be interconnected. Therefore, transmission speed was not a critical factor for survival given the small numbers of connections. As emphasized by David J. Linden in his book "The accidental mind", the brain has never been redesigned from scratch. Therefore, the sluggishness of our neural transmission is a direct result of the absence of re-design. Note that other scientists claim that the existence of delays is a useful feature of the brain and it allows us to compare information over longer periods of time (Nijhawan and Wu, 2009). I believe that there should be other efficient methods than sluggishness to achieve this goal.
These simple examples highlight that our brain is a fortune-teller. It keeps predicting. It infers the features or the motion of objects in our environment, it predicts the consequence of the movement we are going to achieve but also the state of our body (position, velocity of joints, etc...). These predictions are achieved at many different levels: at the periphery, in our sensors (e.g. retina) or in many higher-level brain areas of the cortex. Prediction does not make us human as other species, such as in insects, have this ability as well.
Our brain does not wait to get the information in order to plan and achieve an action. Rather, it predicts what will happen and prepares all the possible relevant actions given its predictions. An action will then be choosen among all the prepared actions (Cisek and Kalaska, 2010). The brain chooses to predict and to plan in advance rather to react to events. It is however, ready to react to any discrepancy between what is actually happening and what has been predicted.
Prediction has driven my scientific work for quite a long time now. In this multiple posts, I would like to highlight that prediction takes place at many different levels and is a inherent feature of the brain.
Here are the topics that will be touched on in the following posts:
- Part II will talk about the predictive mechanisms implemented by retinal cells.
- Part III will describe how prediction drives eye movements
- Part IV will focus on predictive shift of receptive fields in the brain
Pack, C. C., & Born, R. T. (2001). Temporal dynamics of a neural solution to the aperture problem in visual area MT of macaque brain. Nature, 409(6823), 1040-2. doi: 10.1038/35059085.
Zago, M., McIntyre, J., Senot, P., & Lacquaniti, F. (2009). Visuo-motor coordination and internal models for object interception. Experimental brain research, 192(4), 571-604. doi: 10.1007/s00221-008-1691-3.
Orban de Xivry, J., Missal, M., & Lefèvre, P. (2008). A dynamic representation of target motion drives predictive smooth pursuit during target. Journal of Vision, 8, 1-13. doi: 10.1167/8.15.6
Blakemore, S., Wolpert, D., & Frith, C. (2000). Why can't you tickle yourself? Neuroreport, 11(11), R11-6.
Webb, B. (2004). Neural mechanisms for prediction: do insects have forward models? Trends in neurosciences, 27(5), 278-82. doi: 10.1016/j.tins.2004.03.004.
Nijhawan, R., & Wu, S. (2009). Compensating time delays with neural predictions: are predictions sensory or motor? Philosophical transactions. Series A, Mathematical, physical, and engineering sciences, 367(1891), 1063-78. doi: 10.1098/rsta.2008.0270.
Cisek, P., & Kalaska, J. F. (2010). Neural Mechanisms for Interacting with a World Full of Action Choices. Annual review of neuroscience. doi: 10.1146/annurev.neuro.051508.135409.