Recently, I edited a special issue of the journal Interaction Studies on the subject of animal -computer interfaces. The papers centered on dogs and birds, only because two of the promised papers on, respectively, dolphins and apes failed to materialize. (Yes, editing a journal issue or a book is a bit like herding cats.)
The published papers range from discussing some of the history of animal-computer interfaces, to the importance of taking the various animals’ sensory and physical perspectives into account, to a study examining the use of a computer system to interact with dogs. I found the first set of issues particularly intriguing, especially as they might apply to parrots as our companions—are we always doing the best thing for them when we introduce some form of technology into their lives?
The Tech Affect
I want to begin with some of the history, starting with Skinner (1938) and his ‘boxes’ and work on what was labeled ‘operant conditioning’…a way to teach nonhumans to perform certain actions and refrain from others based on the concepts of reward and punishment.
A subject, somewhat food-deprived, would be placed in a box, devoid of anything but a few computer keys; if, for example, it saw a red light and then learned to press the red rather than the green key (i.e., learned “match-to-sample”), it got some pellets; if it pressed the green key, it could hear an unpleasant sound instead.
Notably, Skinner’s techniques and boxes weren’t truly ‘interfaces’ directly designed for the nonhuman subject—that is, something that was innovated for the sole sake of the user/subject, that improved the subject’s well-being, that situated the subject as an eager participant—but rather were mere advances in engineering that enabled humans to efficiently manipulate the subject’s actions to perform a given task and, most importantly, were all for the sake of the human (i.e., to get data).
I am not going to disabuse the importance of such techniques, because they have been adapted and are extremely efficient in training certain basic behavior patterns (e.g., something like getting your bird to ‘climb’ on command, or other simple label-object/action associations). But as I have written extensively elsewhere (Pepperberg, 2021), such training has limitations, as it cannot teach underlying cognitive concepts (e.g., that labels are symbols that can be manipulated and created and used in various non-trained ways for communication); observational learning is far more important for this latter type of instruction. Clearly, scientists—and pet owners—have come a long way since Skinner, based on our knowledge of the physical and psychological abilities and needs of our nonhuman subjects, but several articles in the special issue remind us that we still have a long way to go.
Creating “Soundscapes”
One specific article, by Kleinberger (2023), brings our attention to the sound environment—or the absence thereof—in which various nonhumans, particularly birds, live. She makes us realize that we are often oblivious to what is important for avian well-being. I provide a few examples, with my own comments, below. (BTW…no stranger to animal-computer interfaces, Kleinberger was also involved in designing the ZOOM-like communication system I discussed in an earlier blog.)
She describes a study in which she and colleagues arranged for maternal sounds to be piped into incubators so that birds still in the egg could hear the same sounds as they would if they were raised under their mother in a nest. Other studies have shown that birds who hear their mothers’ vocalizations pre-hatching can, post-hatching, distinguish their mothers’ vocalizations from those of other birds of the same species. These studies introduce a serious issue: When we raise parrot chicks in what are essentially soundproof incubators, are we preventing some important developmental stage (e.g., sound pattern recognition) from occurring? We have no idea, and Kleinberger presents the technology that may help us discover that information.
She also looks at the soundscape to which nonhumans are exposed in places like zoos, but that could have relevance for our companion parrots. Just think for a moment of a zoo you might have visited, where environmental designers decided that having various continuous (often blaring) ‘theme park’ sounds provide a cheery background for patrons, to enhance their mood and encourage them to remain longer, maybe to stay and eat a meal or purchase souvenirs.
A human visitor, however, spends a relatively short time in the zoo. Can you imagine having to listen to such sounds hour after hour, day after day, for your entire life? What if zookeepers decide to pipe in what they believe are natural sounds from the wild to make the experience seem more realistic…but what happens if sounds that are chosen inadvertently contain those of predators or alarm calls that provoke anxiety and fear?
Kleinberger discusses these situations, and I bring them up because I know that some people turn on a radio or TV when leaving their homes, so that the bird will not be ‘alone’…but might such a soundscape really be soothing? Might something come on that causes anxiety or fear? Might the bird spend much of the time calling for their human companions, thinking that they are present? Maybe a well-curated playlist that is constantly being updated—and over which the bird has some control over its presentation—would be more appropriate? Kleinberger discusses a form of this type of technology as well.
Clearly, humans know from their own interactions with technology that it can be a great assistant or present serious problems. We need to extend that knowledge to our use of technology for nonhumans as well!
References
Kleinberger, R. (2023). Sonic enrichment at the zoo. Interaction Studies, 24: 257-288.
Pepperberg, I.M. (2021). A review of the Model/Rival (M/R) technique for training interspecies communication and its use in behavioral research. Animals 11, 2479.
Pepperberg, I.M. (2023). An introduction to “Animal-computer interfaces: Novel approaches for studying animal behavior, cognition and communication” Interaction Studies, 24:193-200.
Skinner, B. F. (1938). The behavior of organisms: An experimental analysis. Appleton-Century.
4 Comments