Secondary somatosensory cortex may be a site of multi-sensory integration and projection
In “The Diverse Mechanisms of Sensory Experience,” I described how each of our senses is encoded in the brain as organized mental maps of sensory space. The story doesn’t end there. For us to make sense of the world, our brain must integrate sensory inputs into a coherent whole. How the brain does this is a major open question in neuroscience.
In a recent review, enticingly titled “Secondary somatosensory cortex of primates: beyond body maps, toward conscious self-in-the-world maps,” Bretas et al. argue for one such site of sensory integration —the secondary somatosensory cortex, or SII. Their ambitious aim in this article was to review SII research in both human and non-human primates in order to understand its functional evolution.
What follows are the big ideas I took away from reading this review, and the questions I was left with.
Linking perception to action: Projection is needed to function in the world
Sensory information travels to the brain where we create a mental model of the information we are receiving. This mental model is called perception. However, the thing we are perceiving is not in the brain, it is outside of us. To act appropriately, we must project our perception onto the source of the sensory information. For example, we look toward the barking dog, and we grab our stubbed toe.
Projection is an ideal mental function that maps (i.e., projects) contents of the internal model of the self and the perceived world onto actual physical environmental worlds.Bretas et al., Experimental Brain Research.
This excerpt is particularly interesting to think about:
Objects in the environment that exist independently of the individual are adopted as references when recognizing one’s own body position in the frame of the external world.Bretas et al., Experimental Brain Research.
Unpacking that a bit, our brain creates a mental model of our body, in part by creating a model of what we can control, and what exists independently (i.e. outside of our control). Breatas et al. point out that our mental models can shift as we gain control over new tools, such as a blind person’s walking stick. Likewise, models shift as we gain control over new body parts, such as a robotic prosthetic, or bodies, such as our avatars in virtual reality environments.
We can also be tricked into projecting sensory information onto a false source. Perhaps the best researched example of this is the rubber hand illusion. A human subject sits with their real hand hidden from view and a rubber hand in sight. About 70% of human beings become convinced that the rubber hand is their own if they see someone stroking it in synchrony with someone stroking their hidden real hand.
Timing is key in creating models of self and world. It is the co-occurrence of multiple sensations that provide the data used for our mental models. As Bretas et al. write, “Multi-sensory integration should be the key mechanism of projection.”
Considering our sense of touch alone, our primary somatosensory cortex (SI) is known to create a mental map of our body parts in relation to each other, but it does not project the body as a whole into a larger environmental context. Bretas et al. write, “To make the projection possible and create the body-in-the-world representation, neural representation at SI must be incorporated with other representations that code information about the external world.”
SII may be a site of multi-sensory integration and projection
SII is peculiar. Neurons in SII respond to large regions of the body being touched. This is in contrast to the neurons of SI, which are organized into a conventional body map —a given neuron is only active in response to being touched along a small region of skin.
Here is where things get interesting. SII becomes active when someone is touched OR when someone observes someone else being touched OR when processing sentences containing textural metaphors. Could SII be important for processing anything related to the concept of touch? The social studies are particularly interesting. Could SII be important for recognizing similarities between self and other, and the development of empathy?
We would like to propose the concept of a somatocentric holistic self, represented as a body-in-the-world map in the primate SII. This concept blurs the border between body and environment by assuming the world as perceived is a product of the self. These mechanisms might comprise the neural underpinnings of the emergence of self-consciousness.Bretas et al., Experimental Brain Research.
Bretas et al. review the latest research finding that SII becomes active in response to many types of sensation, including touch, vision, audition, and vestibular stimuli. This does indeed make it an appealing site for multi-sensory integration. Could SII be important for delineating self from environment, and self from other?
There are no definitive answers yet, but SII seems like a great candidate for further research.
Redundancy breeds functional diversity
Bretas et al. describe how the brain in general and SII in particular expanded in size through evolutionary time, and became quite a bit larger when hominids first started to use stone tools. They suggest that SII was initially redundant to SI, then adapted to have its own function, which became further defined in human beings. They write:
Such mechanisms have been proposed as the theory of neural reuse or recycling (Dehaene and Cohen 2007; Anderson 2010), which claims that neural circuits established for one purpose are exapted (exploited, recycled, redeployed) during evolution or normal development, being put to different uses, often without losing their original functions. The SII of human and non-human primates under this perspective would be an exemplar representation of such process.
One of the most compelling pieces of evidence to support this idea is that SII is highly malleable. In macaques, SII expands in size — up to a 17% increase in grey matter volume — when learning to use a tool for the first time (see Quello et al., where macaques learn to use rakes).
Bretas et al. write:
The proposed neural reuse/recycling hypothesis may have permeated the evolutionarily older brain circuits of SII to accomplish this task while inheriting many of their structural constraints, being yet essentially a center for higher-order somatosensory information processing.
Q1: Does sensory integration occur after individual sensory computations are complete, or simultaneously, in parallel?
Q2: What is needed for sensory integration — sensory hubs where multiple sensory inputs are processed (like SII), brain-wide synchronization, or both?
Q3: How much of the functional diversity of the human brain is dependent upon experience? Are babies born with many redundant circuits that diversify as they learn increasingly complicated tool use, acquire language, etc?
If you have any thoughts on these open questions, please share, either in the comments or by submitting a draft to Awake & Alive Mind.