Potential design ideas (from Ipek):

Thanks Vanessa and Agrita for your comments. How about considering a design where the researchers would ask the owner/human to change something about the environment of the animal, to determine whether the AC’s can detect this change. For instance, researchers could ask the owner to give the animal a new toy,  and have the AC ask the animal about the toy. Or, the owner could be holding a specific object while the AC communicates with the animal. Note that I don’t have that much experience with AC myself, so these are just random ideas for what could be asked to validate that the AC is communicating with the animal & doing so in real time. I am sure members of our group with more AC experience can come up with much better ways to change animals’ environment that will be relevant to the animal and communicable. 

Multiple things would be key with this design: 

1. Only the researcher and the owner would be informed about the nature of the change beforehand, and the AC is not informed to avoid biasing the data.

2. The change in the environment should be something easily detectable by the animal and should be relevant to him/her, so that he/she would be willing to communicate about it. 

3. The change would be kept consistent across different individual animals (e.g. members of the same species), but it would be modified to fit the species that is being communicated with. For example, dogs may prefer different toys or environmental changes than cats (e.g. a new cardboard box may be more exciting to a cat than it is to a dog, for instance).  Similarly, dogs detect colors differently than cats, and so do birds too, so we would account for such species differences when deciding on the change. 

4. Ideal situation would be that we would have the same AC communicate with multiple animals. This would help us understand if the AC is more likely to receive certain types of information and/or respond with certain types of answers when communicating with animals (which is important for data analysis). 

5. If it does not become too much for the animal, we would also ideally have multiple ACs talk with the same animal to check for consistencies. I know that there are some AC courses and some membership programs where animals end up communicating with a whole bunch of different people within a short time frame, so I think this is possible, but it would depend on the personality of the animals of course.