Individual recognition and the ‘face inversion effect’ in medaka fish (Oryzias latipes)

  1. Mu-Yun Wang  Is a corresponding author
  2. Hideaki Takeuchi
  1. The University of Tokyo, Japan
  2. Okayama University, Japan
5 figures


Morphological differences between individual medaka fish.

(A) Medaka individuals may differ in pattern, colour or body shape. The colour and pattern may change based on lighting conditions, physiological conditions and stress level. (B) Mean ± SEM relative reflectance of fish body trunks from five individuals from Figure 1A. Each colour represents one individual fish. Even though the fish look similar under human vision, their reflectance spectra can be very different.
Mating test and electric shock two-alternative forced-choice (TAFC) test were used to examine medaka individual recognition (IR).

(A) Females were familiarised with different types of male cues for more than 5 hr and then the males and females were placed together for mating tests.Grey lines indicate log transformed mean ± SEM time for females to mate. Different letters indicate statistically significant differences after a Tukey’s post hoc test (p<0.05). Each dot represents an individual female. With visual cues alone, the females were able to accept males as familiar mates and required less time to mate. (B) Log transformed time to mate for familiar males (females familiarised with visual cues), unfamiliar males (females given no cue), and exchanged males (females familarised with visual cues from a different male). After substituting the males, the females were able to detect the change and required more time to accept the males. (C) Setup of the electric shock TAFC experiment. The side views of the males were covered. Females were allowed to choose between two unfamiliar males, and when the female entered the area containing the ‘incorrect’ male, she was given an electric shock. When the female remained in the ‘correct’ side for more than 3 min, it was considered that she had made a correct choice, and no shock was given. (D) We tested whether medaka females could discriminate different males with the electric shock-conditioned test. The figure shows the mean ± SEM percentage of correct choices in the electric shock task for two consecutive days. Females were able to distinguish individual males associated with electric shock and performance was improved in the last six trials on the first day. Even after 24 hr, the females could still remember the males and made significantly more correct choices than in the first six trials on the first day.
Illustration of the experimental protocol and the time required for female medaka to mate.

Grey lines indicate log-transformed mean ± SEM time required for females to mate with different groups of visually familiarised males. Letters represent significant differences after analysis of variance tests (ANOVAs) and Tukey’s post hoc tests. Dots indicate individual fish. (A) Female medaka were visually familiarised with a male for different durations. The effect of visual familiarisation was significant after 3 hr of habituation. (B) Pairs of medaka were separated for different durations after being visually familiarised for >5 hr. Even after separation for 3 hr, the females still treated the males as familiar mates; this was no longer the case after 24 hr.
Which morphological traits are important for medaka individual recognition, and how they can be modified.

Grey lines indicate log transformed mean ± SEM time to mate with different groups of visually familiarised males. Dots indicate individual fish and asterisks indicate p<0.05. Letters represent significant differences after ANOVA and Tukey’s post hoc tests. (A) Females were visually familiarised with different types of male visual cues, including appearance and motion. Females were able to recognise the males as familiar mates on the basis of appearance alone. (B) When the head of the male medaka was covered, females were not able to recognise the familiar male and the time to mate was increased. Photos show head-covered and tail-covered medaka. (C) Signals proximate to the head are important for medaka individual recognition. Females were still able to recognise the males after extra spots were painted on the faces of males after visual familiarisation. In the control group, the males were painted by brush with no ink on the face. Photos show one medaka before and after black ink was painted on the face. (D) Images of the males were manipulated with a prism during visual familiarisation, which was followed by mating tests. When familiarised with vertically shifted images, females did not treat the males as familiar mates.
We tested how medaka fish recognise inverted fish and objects.

(A) Setup for the prism glass test. L, left; R, right; D, dorsal; V, ventral. (B) Two sets of non-face-object stimuli were used in the electric shock two-alternative forced-choice (TAFC) tasks. The fish had been exposed to the familiar objects since hatching. (C) Box plots of percentage correct choices from 6 trials before and after the signals were inverted in the TAFC tasks. Fish were trained to discriminate between two fish or two sets of non-face objects for discrete 36 trials, in addition to 6 inverted trials. The ends of the whiskers represent the minimum and maximum of all of the data.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. Mu-Yun Wang
  2. Hideaki Takeuchi
Individual recognition and the ‘face inversion effect’ in medaka fish (Oryzias latipes)
eLife 6:e24728.