Scanning behaviour in ants: an interplay between random-rate processes and oscillators

Abstract

At the start of a journey home or to a foraging site, ants often stop, interrupting their forward movement, turn on the spot a number of times, and fixate in different directions. These scanning bouts are thought to provide visual information for choosing a path to travel. The temporal organization of such scanning bouts has implications about the neural organisation of navigational behaviour. We examined (1) the temporal distribution of the start of such scanning bouts and (2) the dynamics of saccadic body turns and fixations that compose a scanning bout in Australian desert ants, Melophorus bagoti, as they came out of a walled channel onto open field at the start of their homeward journey. Ants were caught when they neared their nest and displaced to different locations to start their journey home again. The observed parameters were mostly similar across familiar and unfamiliar locations. The turning angles of saccadic body turning to the right or left showed some stereotypy, with a peak just under 45°. The direction of such saccades appears to be determined by a slow oscillatory process as described in other insect species. In timing, however, both the distribution of inter-scanning-bout intervals and individual fixation durations showed exponential characteristics, the signature for a random-rate or Poisson process. Neurobiologically, therefore, there must be some process that switches behaviour (starting a scanning bout or ending a fixation) with equal probability at every moment in time. We discuss how chance events in the ant brain that occasionally reach a threshold for triggering such behaviours can generate the results.

Neurons from pre-motor areas to the Mushroom bodies can orchestrate latent visual learning in navigating insects

“…While we acknowledge that ants may sometimes pause for a longer period of time while looking in the direction of their goal, what is clear is that learning walks are rather optimised to collect a diverse sample of views in all directions. This is in line with previous works showing that pauses and scans are not tightly controlled by the relation with the ant and its environment, but rather are the result of 'blind' internal motor processes such as the continuous production of regular oscillations in the ant's angular and forward speed 22,49 as well as the random triggering of pauses 42 . This stochasticity is further highlighted by the great variability in the expression of learning walks observed across individuals 50 .…”

“…The latter requires computational steps, such as Fourier transforms, which potential implementation in insect circuits remains unknown. Instead, assuming an egocentric encoding enables to remain faithful to the known insect neural circuits and corroborates the behavioural evidence that ants' visual scene recognition requires the insect to align its gaze as during training 66,[71][72][73] as well as the regular need of insects to scan multiple directions 21,42,49,74,75 .…”

“…Artificially restraining these exploratory movements (in both time and space) reduces the ants' subsequent navigational performance based on terrestrial cues, showing that they do learn the scenery during these exploratory behaviours 39,41 . At a finer scale, these meandering trajectories are interspaced with regular slowing down up to complete halts, producing behaviours so-called votes or scans, whose expression varies across species and individuals 42,43 . Slowing down and pausing helps the ants obtain a stable (and thus non-blurry) view and therefore surely contributes to visual learning.…”

Unraveling the neural basis of spatial orientation in arthropods

“…These movements likely aid in panoramic view matching and deciding which path to follow. Sudhakar Deeti and coworkers have analyzed these saccades in detail to derive hypotheses about how they may be generated in the brain (Deeti et al 2023 ). They show that this behavior has a particular temporal structure suggesting an underlying random-rate process.…”

Nest excavators’ learning walks in the Australian desert ant Melophorus bagoti

“…Statistical analyses were conducted using R (version 4.3.1). During learning walks, ants frequently displayed a series of stereotypical successive fixations in different head directions by rotating on the spot at one location, known as a “scanning bout” and we extracted the number of scanning bouts from each learning walk (Deeti et al 2023c ; Lionetti et al 2023 ).…”