Reliability-based cue combination and sensory adaptation theories

Reshaping the Sensory Environment

Sensory Accuracy

Survival of all animals depends on the accuracy with which sensory information is processed by the nervous system. Integrating this information in an efficient and effective manner depends on dynamic strategies that the nervous system relies on to determine the reliability and accuracy of the sensory inputs (reviewed by Zaidel, Turner, and Angelaki, 2011). This essay examines contemporary theories that attempt to explain how these strategies function.

Reliability-Based Cue Combination (RBCC) Theory

The reliability of a sensory cue is believed to determine the weight an organism assigns to a given cue, such that a more reliable cue may have a greater influence on behavioral outcomes (reviewed by Zaidel, Turner, and Angelaki, 2011). Empirical support for this theory has come from studies examining the integration of multisensory information. This theory has been called reliability-based cue combination (RBCC).

Reliability, as defined by RBCC, is used interchangeably for sensory precision; however, these terms do not necessarily imply accuracy (reviewed by Zaidel, Turner, and Angelaki, 2011). Accuracy is defined as the degree of agreement between the sensory information and the environment. External feedback is theoretically required for determining accuracy, but in the absence of additional external information, the organism must make do with the degree of agreement between different sensory inputs. For example, the sound of an approaching car can be confirmed visually, but should someone experience hearing a loud bang outside their visual field, an internal validation process will likely occur because visual confirmation is impossible.

Reliability-Based Adaptation (RBA) Theory

In situations when multisensory inputs disagree, RBCC proposes that the sensory cue with the most reliability will have the greatest influence on behavior (Burges, Girshick, and Banks, 2010, p. 8). Accordingly, if visual information is providing the most reliable cues for a given environment, discrepant auditory or touch information will be discounted or undergo adaptation and/or recalibration relative to the visual information. This theory has been called reliability-based adaptation (RBA) (reviewed by Zaidel, Turner, and Angelaki, 2011). Although some researchers have suggested humans are visually dominant and therefore all other senses adapt to visual cues, recent research provides strong support for the theory that all senses are capable of adapting depending on the situation (Burges, Girshick, and Banks, 2010). For example, hearing a nearby train whistle at night may cause someone to hesitate before crossing a set of railroad tracks, even if they cannot see an approaching train. In this environment, the reliability of aural sensory information is increased relative to visual information because it is nighttime.

Despite the empirical evidence indicating the visual system can adapt to a more reliable sensory cue, other studies have provided evidence suggesting that visual information probably represents a default sensory reference in most situations where visual cues are provided. When barn owls had prisms placed in front of their eyes their prey-oriented flight paths suffered, despite the use of both hearing and vision to find and capture prey (reviewed in Burges, Girshick, and Banks, 2010, p. 10). With time however, the owls adapted to the prisms and the accuracy of their flight patterns increased. When their brains were compared to control animals the connections in the auditory cortex, rather than the visual cortex, were modified. Visual dominance, in the form of neurobiological correlates, therefore seems to exist in barn owls.

Discussion

Sensory information from multiple sources, such as vision, hearing, and touch, are managed by the nervous system to produce an accurate representation of environmental conditions. Because of the complexity inherent in such a system, there is considerable potential for error. When error does occur, the brain prioritizes sensory cues depending on how reliable this cue has been in the past in providing useful information. If external feedback is possible, sensory inputs can be recalibrated over time to increase the accuracy, and thus reliability, of the cues. Judging the accuracy of a sensory input therefore depends on how much agreement there is between multiple sensory inputs, how reliable these sensory inputs have been in the past in providing useful information for navigating the environment, and the novelty of the environmental situation.

The accuracy of a sensory cue can depend on many factors. If limb movement is used as an example, then vision, perceived effort, muscle stretch receptors, and position relative to gravity, combine to provide the sensory information needed to determine position, speed, and the trajectory of the moving limb. If a limb is artificially supported, pre-conditioned, or fatigued, then accuracy decreases when a person, who is deprived of their vision, is required to determine limb position (Winter, Allen, and Proske, 2005). However, the expectation would be that the person would be able to adapt over time and increase sensory accuracy using external feedback. Some of the factors contributing to sensory accuracy are therefore nervous system health and an experiential history within a defined environment. Another factor that can decrease sensory accuracy, which was not discussed above, is the amount of sensory noise inherent to an environment. Too little or too much information can impair or preclude sensory information processing. For example, relying too heavily on visual information at night may not be the best strategy for avoiding being run over by a train.

The empirical evidence presented here reveals how plastic the nervous system can be with respect to processing sensory information. While the existence of a functional nervous system depends almost completely on genetic determinants (nature), it can be shaped on a neurobiological level in dramatic ways by sensory experience (nurture). This was revealed in a dramatic way by the barn owl experiments. The processes of sensory cue calibration may involve only cognitive adaptation in the short-term, but adaptation over a prolonged period probably relies heavily on rewiring the brain. However, the prevalence of phantom limb sensations in amputees suggests that extinction of some neuronal pathways may not be possible using sensory recalibration and adaptation.