Synonyms

ACI; Animal-computer interface; DCI; Dog technology; Dogs

Definition

Animal-computer interaction (ACI) and dog-computer interaction (DCI) are extensions of human-computer interaction (HCI), which refer to the usage of technologies for animals and dogs, respectively. However, HCI refers to usage of technologies for humans.

Introduction

Animal-computer interaction (ACI) includes developing a video game prototype for canines. This entry aims to provide context for the problem of attracting dogs to watch screens; the solutions that have been presented, with aims, objectives, and functional/nonfunctional requirements; the perspective that this project makes to the research field of ACI; and the scope of the entire project (Mancini 2016).

ACI is a relatively young but expanding field of research with a lot of progress made in recent years all across the world. However, there are still a number of questions that need answering, such as the following: How do we change the design of human applications to allow animals to interact with them (Mancini et al. 2012; Mancini 2013, 2017)? How can we have animals use technology without human assistance (Hirskyj-Douglas and Read 2014)? What kind of applications benefit animal welfare (Geurtsen et al. 2015; Broom 1996; Rehn and Keeling 2011)? How can we analyze how dogs feel about the applications that we present them with (Baskin et al. 2015; Mankoff et al. 2005).

All of these questions are being asked so that researchers can understand how to properly design various hardware and software for future ACI projects, particularly for dogs. However, although these questions are important in the design of the research, the main question that authors are trying to solve with it is: What can be done to attract dogs to a computer screen in order to use the application created for them?

One of the most essential steps in ACI design is getting the dogs to initiate interaction with the application; however, in order to do that, they need to be interested in interacting with the application in the first place. To achieve this, developers of ACI applications need to establish appropriate techniques that they should be employing in the design of their applications. The problem, however, is that there are no definitive ways of implementing this essential step in ACI design. This leads to extra time and resources being spent by developers trying to understand what exactly they can do in order to solve this problem themselves. A comprehensive review is needed in detail of the various solutions which researchers in the field have produced, to see which ones can be generally applied to each type of application, based on the nature of the hardware and software that are need for the developing stage.

To design and implement any interactive application for animals (Zeagler et al. 2014) in the form of a video game prototype that attracts the attention of dogs and be useable by them (Nielsen 1995; Kjeldskov and Graham 2003), some main steps are required which will be discussed in this entry. This means designing the application using various UML diagrams and techniques; developing a demo via chosen programming language; and having a selection of test participants (all canine) to interact with it, under ethical conditions (Grillaert and Camenzind 2016), in order to ensure that the application meets the specified requirements. The most important aspect usually is to ensure that the video game is fit for purpose. This is the most crucial part of the study, as cannot be solved the main problem of understanding what attracts canines to screens without ensuring that it is certainly a video game with which dogs would be interested in interacting. Otherwise, it would be pointless to have controlled experiments, as they are more than likely to ignore anything that is not designed with them in mind. This research references to what dogs are interested in and looks to incorporate those interests in the video game, in order to come up with the list of interests for dogs in the field of ACI.

Methods

Giving the canine incentive to notice and pay attention to the screen which your application is being displayed on is one of the most essential steps in initiating interaction between dogs and software. This is evident in the experiments that has been researched which all used some technique to attract dogs to their own screens (BBC 2018; Baskin et al. 2015; Geurtsen et al. 2015). The use of techniques like this can also provide additional benefits to familiarize dogs with the hardware that researchers want them to use, such as the experiment by Dr. Lisa Wallace whose use of screen paste taught dogs how to interact with their own touchscreens using their tongues. Various such techniques will be documented, and their effectiveness will be detailed within this entry, so that researchers can understand why they are important and possibly use them in their own experiments (Burghardt 2005).

Placing Confectionery Items on a Screen

One of the more unique techniques for attracting dogs to a screen that researchers and scientists have been using as of late is to place confectionery items on the screen of the computer/tablet. Dr. Lisa Wallace from the Eötvös Loránd University has been using this technique in order to try and get dogs to interact with their brain training program on a tablet touchscreen. They would first smear a flavored paste all over the touchscreen without turning on the tablet and have the canines lick it off. They would then start the application and smear some more paste on the screen to have the dogs make contact with the application using their tongues. Once the application detected them touching the screen, they would receive a dry treat from a dispenser below the tablet; after some time, paste was no longer required to convince the dogs to lick the touchscreen and interact with the application, as they had now learned how to interact with it as the scientists had intended (BBC 2018).

Sound Design

One of the best senses that dogs possess is their ability to hear, which is generally considered better than that of humans. Their hearing spectrums range from 40 hz to 60 khz (Jensen 2007), and moreover they can locate the source of distinctive sounds (Geurtsen 2014). As stated earlier, we want dogs to exhibit playful behavior and hunting instincts with the prototype, and thanks to this excellent hearing range, certain sounds can be used to help initiate this behavior when it is needed to interact with the video game. The sounds that dogs will react to can vary on a case by case basis: some may prefer the sounds generated by squeaky toys if they particularly enjoy playing with them, or if a dog prefers hunting small animals, then it would probably respond to sounds that they would expect to hear from small animals, e.g., mice rustling through grass (Singer 1995).

A case has been made for this by researchers Golbeck and Neustaedter (2012), who in their project to design a pet monitoring device (for pet owners and canines to call and interact with each other) used a soundboard (among other ideas) to attempt to attract dogs to their prototype system. The soundboard consisted of sounds such as dog barks/howls, cat sounds, squeaky toys, and other sounds that they believed would interest dogs and could be activated remotely by the pet’s owner to play the sounds through the monitoring system. In their experiment to have dogs interact with their monitoring system, they would have human users use the soundboard to attract their dogs to a pre-placed laptop screen in order to start interacting with their pet through the monitoring system. All but one of the dogs successfully came to the screen, the lone exception being a golden retriever that was too excited to pay attention to the sounds being played (Golbeck and Neustaedter 2012). One of the best ways to recognize what sounds dogs would want from an application would be the use of dog personas gathered from canine participants (Hirskyj-Douglas et al. 2017).

Laser Pointer and Visual Cues

The same experiment mentioned above also attempted to make use of the dog’s visual sense by two different means. The first was a laser pointer, and the second a virtual object in the form of a tadpole, both of which were displayed remotely on the screen.

Discussion

This section contains discussions on the effectiveness of each technique and what have been considered in existing methods to make them engaged as much as possible.

Placing Confectionery Items on a Screen

Placing confectionery items on a screen has proven to be very useful for the particular experiment demonstrated as they managed to show how useful it was for training dogs to use their devices with their tongues. Eventually they managed to have their dogs use the tablet device without the need for confectionary pastes on the screen, proving that this is indeed a great technique for having dogs engage with a computer screen.

There is only one problem with this technique, which is that it really requires specific equipment in order for anyone to make use of it properly. In the experiment, it was specified that the confectionery items they placed on their tablets was a flavored paste that was completely smothered all over the screen for the dogs to lick off. The problem with this is that most computers are not designed to have any kind of confectionery item, let alone flavored paste, smothered over their displays. There should be a few exceptions such as the devices they used in their experiment. (Note: The devices used in the experiment were not specified in the information source that have been examined (BBC 2018).) However, it would require dog owners to go out of their way to find that equipment if they want to try this technique themselves. Most dog owners would be completely averse to trying this technique, so as not to risk causing any kind of damage to their own computers due to the use of paste on their displays. This particular technique has not been taken into account as there was no access to the necessary technology, but there are rooms to acquiring the means necessary to try it in the future.

Sound Design

The use of specific sounds in DCI seems to be the most viable and generally best method for attracting dogs to a screen. This is for several reasons, the first of which is that dogs very much respond to recognizable sound effects and will tend to locate the source of recognizable sounds as soon as they hear them; therefore it should theoretically lead them to the computer, which they will investigate upon realizing that it is the source of the sound. Again, this has been proven by the experiment by Jennifer Golbeck who had almost all the participants in their experiment attracted to the screen via their own pre-developed soundboard (Golbeck and Neustaedter 2012).

Another reason for this being a viable technique is that it is a very easy method to use for attracting dogs to the screen, since all you really need is a variety of sound files that would attract the attention of canines upon hearing them. Developers of dog-centered applications and dog technology should be able to make use of this technique in some form by implementing sound files within their projects.

The only problem that some people would have with this method is that developers would have to take time to research what kind of sounds dogs would be interested in. This is to ensure that they use as many as possible within their projects, as some dogs may not have the same interest in certain sounds as others. Dog owners would need to experiment themselves and see which sounds from the variety that developers have provided catches the attention of their particular canine.

In the experiments with this technique, the majority of the participants expressed a moderate to great amount of interest toward the computer when sound effects are used in the video game prototype, particularly squeaky toy sounds. Some of their behavior patterns indicated interest toward the source of the sound as they would bark and jump toward the screen recognizing it as the source of the sound. However, the only problem is that during the experiments, one of the canines was confused by the sounds, as they would attempt to stick their nose and paws underneath the stool that the host device was placed on. This indicated that the canine did not recognize the laptop as the source of the sound and instead was looking for a squeaky toy underneath the stool.

Laser Pointers and Visual Cues

Judging by the research, this technique may not be as viable as the other methods described earlier, as proven by the experiment by Jenifer Golbeck et al. In their own experiment, they tried to attract the attention of dogs with both a laser pointer and a tadpole game playing on their computer screen, but with little success of attracting the dogs successfully to the screen (Golbeck and Neustaedter 2012). This is because these objects were simply not very interesting to the canine participants, as other experiments have managed to successfully attract dogs to screen via screen cues, such as the one done by Sofia Baskin et al. Their experiment used two different kinds of visual cues, both a ball and mouse type object that would move around the screen. It was reported in this entry that both of the canine participants used in this experiment had taken interest in the objects by exhibiting certain behavioral patterns that indicated this (Baskin et al. 2015). That leads me to conclude that the use of visual cues can only work on a case-by-case basis for canines as they are only individually interested in certain objects.

If developers want to use visual cues, then it would be wise to implement a variety of objects in their projects to ensure every dog they work with has something that will interest them. However, depending on what kind of application they are developing, they will either have to develop these cues themselves, which take a lot of time and work, or get them from somewhere else, such as the Unity Asset Store. The latter option can be considered very expensive, depending on what kind of assets the developer needs, how many they need, and whether they are made at the quality level that the developer needs them to be at in order to use them properly.

Conclusion

When comparing all three options for attracting canines to a screen, it is safe to say that the most viable method that developers can utilize is the use of sound cues. This is due to various facts including that any sounds that developers require are easy to access and, in most cases, do not have any drawbacks to downloading them, such as fees; they are very easy to implement in most software applications aimed toward dogs; and they have been proven to be the most successful method when tested with a large range of dogs (Golbeck and Neustaedter 2012).

However, this does not mean that the other two methods are nonviable, as they each have their own useful factors for attracting dogs to a screen. Factors such as teaching fogs on how to use a touchscreen with their tongues, as well as to attract them to the screen, and keeping them engaged with the screen for a prolonged period of time when they notice that something of interest is there have been investigated by Baskin et al. (2015).

Overall, the viability of each method depends largely on the kind of hardware and software that DCI developers are using for their particular project. The viability also depends on the variety of assets used with each method and how well they use them for getting the dogs’ attention. If the available options for each method are researched well, chosen well, and implemented to the highest standard expected of them, then they should overall have success in attracting dogs to their computer screen.

Cross-References