No statistical significance was found that during testing, the macaws performed differently from chance level

I hypothesized that by creating a situation where the macaws should have a very strong expectation to copy the behavior of a conspecific, presenting the subjects with an unexpected repeat command would require the macaw to recall its own actions through episodic-like memory. Similar studies have assessed this ability in other animals such as dogs and pigeons, obtaining around 60% of correct responses upon receiving an unexpected test, a performance above chance level in both cases. Further studies are recommended to assess whether the current results are due to the low statistical power, or if this capacity is truly not present in blue-throated macaws.

Due to the low amount of testing trials per individual, analyses were done at group level only. When there is a need for the test to be unexpected, repeating the test could result in it being anticipated, hence the low amount performed in the study. However, individual differences may have influenced performance, such as the inclusion of more complex behaviors, as three out of five behaviors were randomly selected for each individual. Therefore, testing each macaw enough times for individual analyses to be performed could have shed more light into this topic. Alternatively, A bigger sample size could dilute the effects caused by possible individual differences.

Macaws looked at the hand commands significantly longer when those were unexpected

The presence of this effect brings evidence that the test used in the current study was unexpected. This could serve as a control measure to assess if the test remained unexpected as more testing sessions were performed – allowing future studies employing the same paradigm to perform more testing sessions per individual while measuring whether the test was still unexpected, bringing more statistical power to it.

These findings are congruent to what has been reported as an effect of this paradigm in different mammalian species:

  • Human infants look reliably longer at an impossible event when in comparison with a possible event (Baillargeon, 1987).
  • Dogs gaze longer at the owner when receiving an unexpected imitation test than in a baseline condition where the test was expected (Fugazza et al., 2016).
  • Belugas gaze longer at unexpected experiences than at baseline conditions  and expected experiences (Hill et al., 2023).

Knowing that it is also an observable effect in macaws opens the door for future macaw studies to make use of this paradigm, possibly also with other parrot species.