An Action Field Theory of Peripersonal Space

Abstract

Predominant conceptual frameworks often describe peripersonal space (PPS) as a single, distance-based, in-or-out zone within which stimuli elicit enhanced neural and behavioural responses. Here we argue that this intuitive framework is contradicted by neurophysiological and behavioural data. First, PPS-related measures are not binary, but graded with proximity. Second, they are strongly influenced by factors other than proximity, such as walking, tool use, stimulus valence, and social cues. Third, many different PPS-related responses exist, and each can be used to describe a different space. Here, we reconceptualise PPS as a set of graded fields describing behavioural relevance of actions aiming to create or avoid contact between objects and the body. This reconceptualisation incorporates PPS into mainstream theories of action selection and behaviour.

Keywords: action selection; defence; egocentric coding; goal-oriented actions; motor system; perception and action.

Figure 1

Describing Peripersonal Space (PPS): Gradient or Boundary? Many behavioural and neurophysiological responses have been labelled as PPS measures because their magnitude increases with body proximity. However, even when clearly graded with proximity to a body part, PPS measures are often described using binary ‘in-or-out’ metrics and wording (left side of each panel). This approach often consists in choosing some cut-off value to define the PPS ‘size’. Examples of such cut-off values are the furthest distance at which consistent modulation is observed (A) and the midpoint of a fitted function (B,C). (A–C) show examples of how PPS data could be described as binary ‘in-or-out’ metrics [left side of (A–C)], but more faithfully reflect the data when displayed as a continuous, graded response field [right side of (A–C)]. (A) Bimodal visuotactile neurons fire more when visual stimuli are close to their tactile receptive fields . However, as discussed in the main text, this is an oversimplified description. (B) The hand-blink reflex is elicited by stimulation of the median nerve at the wrist, and increases in magnitude when the hand is closer to the face . (C) Reaction times (RT) to somatosensory stimuli on the face (green), hand (blue), and chest (red) are faster when auditory or visual stimuli are concomitantly presented closer to those body parts . Other types of visuotactile and audiotactile integration have also been shown to increase in magnitude with proximity the body . Data reproduced, with permission, from , , .

Figure 2

Many Peripersonal Space (PPS) Fields, Affected by Many Factors. As detailed in the main text, we propose a reconceptualisation of PPS as a set of fields reflecting the relevance of actions aimed at creating or avoiding contact between objects and the body. This figure illustrates the idea that there is not a single PPS, but that instead there are many PPS fields. (A) Heterogeneity of PPS fields. Different PPS fields can be derived from the many types of biological measure that differently depend on spatial proximity. Here, we show as an example the PPS fields derived from the modulation exerted by the proximity between a visual stimulus and the body on different biological measures: (i) the somatosensory-evoked eye blinking (green); (ii) the response of a visuotactile single neuron with a somatosensory receptive field on the chest (red); and (iii) the reaction times (RT) to somatosensory stimuli delivered to the hand (blue). Note how the same visual stimulus in an identical position elicits different responses and defines PPS fields with different spatial features. (B) Not just proximity: additional factors modulate PPS-related measures. Although the magnitude of PPS-related measures is commonly affected by proximity to a body part, many other factors also affect these PPS measures. Such factors include various types of motion: motion of a visual stimulus (i), stimulated limb (ii), and the entire body (iii) can all cause expansion of the response fields. Factors independent of motion also affect PPS measures: tool use can expand response fields (iv), a protective screen can deform them (v), and frightening sounds can expand them (vi). Response fields are colour-coded by the body part near which their magnitude is maximal: face (green), hand (blue), and trunk (red). Data reproduced, with permission, from , , , , , , .

Figure 3

Peripersonal Space (PPS) as Action-Relevance Fields. We conceptualise PPS as a set of fields reflecting the relevance of actions aimed at creating or avoiding contact between objects and the body. We borrow the term ‘field’ from modern physics, to express a quantity that has a magnitude for each point in space and time . (A) Magnetic field strength decreases proportionally to the cube of the distance from a coil. Therefore, when approaching a magnet holding a metallic object, one perceives a ‘boundary’ where the attractive force becomes detectable. Still, the magnetic field covers the entirety of space, although its strength becomes infinitesimal. (B) Similarly, PPS field values can be infinitesimal in most of the space far from the body, although this does not mean that they have zero value. Admittedly, considering such distant PPS field values is rarely phenomenologically useful, but it does become important in those instances of PPS measures with either extremely large response fields or spatially separate regions with strong responses . This panel shows the PPS field for the defensive eye blink . Envisioning PPS as a set of response fields avoids the oversimplification of considering it a single and binary ‘in-or-out’ bubble, and it allows a richer description of the actual response properties in space. (C) The notion of contact-related action fields fits well with the interactive behaviour framework . This framework, exemplified here for visually guided movements, postulates that neural populations in the parieto-premotor loop transform visual input into representations of potential actions. The strength of each action representation is also influenced by basal ganglia and prefrontal regions so that it reflects the relevance of that action depending on the environmental circumstances. Reproduced, with permission, from (C).

Figure 4

Effect of Context on Peripersonal Space (PPS) Fields. Consider two possible actions when facing a predator such as a crocodile: climbing a tree (red) or running away (blue). Given that the relevance of these actions depends on the position of the predator (second row), any measure reflecting the relevance of these actions can be used to map out a response field in space (third and fourth rows). Examples of these measures might be the probability of observing the given action, the firing rate of a neuron that is involved in preparing the action, or the reaction time to a sudden tactile stimulus on the body. In this perspective, the fields described by the magnitude of these measures are instances of PPS fields: fields that reflect the relevance of actions aimed at avoiding or creating contact. This conceptualisation (i) allows PPS measures to change gradually with distance; (ii) reflects the fact that many different PPS measures show different response profiles; and (iii) explains the functional significance of the values comprising the PPS field of each action, and the fact that factors other than proximity affect PPS measures. This example encapsulates those three points. In the presence of a crocodile and a tree (left column), climbing is an action the relevance of which increases with the proximity between the crocodile and the individual, unless the crocodile is interposed between the two (red). In the same situation, running away (blue) is a less beneficial action unless the crocodile is interposed between the tree and the individual. By contrast, in the absence of a tree (right column), climbing becomes irrelevant, while running away becomes more relevant.