(C) PLOS One
This story was originally published by PLOS One and is unaltered.
 . . . . . . . . . .



Eurasian jays (Garrulus glandarius) show episodic-like memory through the incidental encoding of information [1]

['James R. Davies', 'Department Of Psychology', 'University Of Cambridge', 'Cambridge', 'United Kingdom', 'Elias Garcia-Pelegrin', 'National University Of Singapore', 'Singapore', 'Nicola S. Clayton']

Date: 2024-06 

Seven Eurasian jays (3 females, 4 males) participated in this study. The sample was made up of all birds that were available and motivated for testing. All the jays were hand-reared in 2015 and were socially housed within a large outdoor aviary (approximately 20m long × 10m wide × 3m high) at the Sub-Department of Animal Behaviour, University of Cambridge, Cambridgeshire, UK. The aviary was divided at one end into smaller sections (approximately 6 × 2 × 3m) which connected to indoor testing compartments (each 2 x 1 x 2m). The jays were fed a maintenance diet of soaked cat food biscuits, eggs, vegetables, seeds, and fresh fruit and had ad libitum access to water (including during testing). To increase their motivation to participate in experiments, the jays’ maintenance diet was removed from the aviary 1-hour before testing. Subjects participated on a voluntary basis (maximising motivation) and were individually separated once they entered the testing compartments. The experimenter interacted with the jays via an open window adjacent to the indoor compartments.

Procedures

The experiments were reviewed and approved by the University of Cambridge Animal Welfare Ethical Review Body and were conducted under a non-regulated procedure license (NR2021/49). All procedures were non-invasive, purely behavioural and did not require anaesthesia or euthanasia of any subjects. The jays had previous training using cups in previous studies (e.g., [51]). The cups in this study, and other unpublished studies, were always identical and plain red with no other salient and distinctive visual features, as the cups in these experiments simply represented spatial locations that food could be hidden under. Therefore, the jays had no experience with being trained to attend to any visual features (unique or not) on the cups before the onset of this experiment. To confirm their ability to locate food hidden under the cups, a spatial memory training phase was conducted. In this phase, the testing compartment was set up with a platform next to the experimenter window, with a perch placed at the centre. Four identical cups, equally spaced, were arranged in front the perch, parallel to the experimenter window (Fig 1A). The bird entered the compartment and was separated from the rest of the aviary; now only having access to the test compartment and an adjacent closed outside section of the aviary. Whilst remaining on the perch and facing the experimenter, the jays observed as the experimenter turned one of the cups over, placed a mealworm inside, then returned it to its original position (Fig 1A). The birds then made a choice by pulling on a string attached to the top of the cup (S1 Fig) to reveal its contents (Fig 1B). Once a choice was made the trial ended, and the bird was encouraged to return to the perch before the procedure was repeated for a subsequent trial. After baiting a cup, the experimenter orientated their head and eyes directly forward and kept their arms by their side in order to limit the possibility of accidental cues directing the bird’s choice. The position of the baited cup was pseudorandomised, in that no single cup contained the mealworm more than twice in a row. All birds reached training criterion (8/10 successful trials) in a single session of 10 trials.

The set up and procedure of the test phase was almost identical to the training phase, except for the cups used and the addition of an unexpected delayed long-term memory assessment. Instead of identical cups, the cups used in test trials each possessed a unique visual marker as part of an array of different visual elements: either a) coloured card around the string attached to the top of the cup (S1A Fig); b) a laminated coloured shape attached to the front of the cup (S1B Fig); or c) a laminated coloured and/or patterned card underneath the cup (S1C Fig). These visual characteristics were not present during training, and thus did not represent relevant information associated with solving the original trained task. Each bird received a single trial per trial type (string, shape, or card), each on different days separated by at least 24 hours. The order of trials each bird received was pseudorandomised (so that the position of each trial type in the sequence was counterbalanced across individuals) and unique (except Stuka and Booster; S1 Table). One bird, Sojka, failed to make a choice in her string trial within 15 minutes after entering the test compartment and so was excluded from this trial type. Immediately before each test trial, a minimum of 5 and a maximum of 10 trials (identical to the training procedure but with the test cups) were conducted to ensure the jays were still performing accurately with spatial memory and not failing due to other factors unrelated to the study (e.g., fear, lack of motivation, etc.). As no cup was baited more than the others, and the bird had immediate access to recover the reward underneath the cup, the visual characteristics of the cups were again not relevant to solve the task at this stage. As with the training phase, the only information necessary to solve this task was spatial information, meaning the birds only had to use short-term spatial working memory to be successful. Once an individual reached 5 consecutive correct re-training trials, a single test trial was conducted. The first stage of this trial (Fig 1i) was identical to the training procedure, except that 5 mealworms were used to bait the cup (to increase the saliency of the caching event) and the bird was prevented from reaching the cups to prematurely make a choice (a plastic bar was placed over the cups). Once the cup was baited, the bird was removed from the testing compartment (into the outdoor section) and was left alone for 10 minutes without visual access to the cups or the testing compartment. Concurrently, the experimenter moved the cups into a new, distinct location (perpendicular to the test window) and positioned them at random (Fig 1ii). The mealworms were removed from under the cups to control for any visual, auditory, or olfactory cues, thus forcing the birds to rely on memory alone in the succeeding test. Once 10 minutes passed, the bird was brought back into the test compartment and allowed to make a choice. The compartment doors were set up so that the birds faced the cups from the front, and thus were roughly at an equal distance from the jay upon their presentation.

As the visual markers were present on the cups during the re-training trials before each test trial, the birds could potentially use two different strategies to solve this task: 1) rely on the spatial information, or 2) rely on the visual information (or a combination of both strategies). Whilst the birds were exclusively trained to use spatial information to solve the training task, although unlikely, the use of the second strategy cannot be ruled out using this methodology alone. Therefore, an additional control task was conducted to determine whether the birds prioritised spatial information over visual and whether they would learn to use visual information when spatial information was no longer relevant, across the same number of trials as the retraining minimum (n = 5). This task was identical to the re-training trials, except that before the bird was allowed to make a choice, the cups were quickly rearranged out of sight (behind a visual occluder). This way, the original spatial cues were still present (i.e., there was a cup in the same location as before) but now the use of the first strategy would lead to failure. The position of each cup after rearranging was pseudorandomised so that the baited cup was never in the same spatial position as it was previously. Crucially, if the jay chose the cup with the same visual marker (rather than in the same position), they received the reward underneath it, thus allowing them the opportunity to learn to solve the task using visual information. Each of the 7 jays performed this task following the main testing period, and the visual markers used (trial type) were counterbalanced across subjects (with an extra session of ‘shape’).

[END]
---
[1] Url: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0301298

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/