What did we want to find out?
Typically, studies of successive negative contrast are conducted using food rewards of varying quality or quantity (as discussed here). There is evidence to suggest that many animals are able to show behaviours akin to disappointment when there is a sudden downshift in reward quality or quantity of food or liquid rewards, but to date this has not been investigated in relation to social rewards.
There are several studies that have shown the same areas of the brain to be responsible for processing social and non-social rewards, and that social rewards themselves can act as a natural reward to many species (shown by the fact that animals are prepared to work for contact with conspecifics).
As highly social animals that form close bonds not only with conspecifics but also with people (e.g. Topál et al. 1998), domestic dogs are well suited to investigate this question. Dogs generally show a preference for their owner over an unfamiliar person – indicated by seeking contact with their owner more than with a stranger, waiting by a door when their owner has left the room, and greeting their owner more enthusiastically than greeting a stranger.
Therefore, the aim of this study was to investigate whether dogs were capable of showing behaviours akin to disappointment when, upon them expecting their owner to enter a room, an unfamiliar person entered instead.
How did we do this?
17 dogs of various breeds (9 neutered females, 7 neutered males, one intact male), ranging in age from 12-132 months, were enrolled in the study, which was designed as a within-subjects study (meaning that each animal experienced both the unshift and the downshift conditions over 2 consecutive weeks).
Before the study started, dogs were given time to explore the room and interact with their owner and the unfamiliar person. A third person (the observer) was also present but did not interact with the dog at any point during this phase. After the familiarisation period, the unfamiliar person and the owner left the room, leaving the dog alone in the room with the observer, who ignored the dog, and the trials began.
Trials were conducted sequentially, without breaks, and each trial consisted of three stages (Table 1):
Table 1: Description of the three stages
|Stage 1 (anticipation of reward)||Dog hears a knock on the door, and after a period of time the owner or unfamiliar person (dependant on condition) enters the room.|
|Stage 2 (access to reward)||The owner/unfamiliar person is seated on a chair in the room for 30 seconds.|
|Stage 3 (loss of reward)||The unfamiliar person/ owner exits the room. Stage 3 begins from when they have left the room until the knock on the door sounds again, signalling the start of the next trial.|
In the unshift condition, an unfamiliar person entered the room in every trial. In the downshift condition, the dog’s owner entered the room in the initial trials, followed by a “downshift” in reward to the unfamiliar person entering the room, concluded with trials where the owner entered the room again.
When the owner/unfamiliar person were seated on the chair in the room with the dog, the dog was able to initiate tactile interaction with the person, but was also free to choose not to do so.
Dog greeting and interacting with owner
Dog greeting and interacting with unfamiliar person (when expecting owner to enter)
Dog greeting and interacting with owner (after interacting with unfamiliar person)
What did we find?
We found that dogs behaved very differently when their owner was present compared to when the unfamiliar person was in the room with them. They spent more time focussed and in contact with the owner compared to the unfamiliar person, and they spent less time gazing at the door when with the owner than when with the unfamiliar person, indicating a clear preference for the owner over the unfamiliar person. Most interestingly, during the downshift condition, when dogs were expecting their owner to enter the room and instead it was the unfamiliar person, they spent significantly longer looking at the door compared to when the unfamiliar person entered during the unshift condition.
This is consistent with findings of SNC in consumatory studies: when an animal experiences a downshift in the quality or quantity of a food reward they appear to be “disappointed” and spend time searching for a better reward.