ConductScience VR
Conditioned place preference in virtual reality
People visit two virtual rooms, one paired with a reward, then choose where to spend their time, as rats and mice do in a place preference box.
ConductScience Editorial · Published November 15, 2026 · 5 references
- Task
- Spend time in two virtual rooms, one paired with a reward, then move freely between them with the reward gone.1
- First shown
- Molet and colleagues published the first virtual reality place preference study in 2013, with music as the reward.1,2
- Measures
- Time spent in each room, and ratings of how much people like each one.1
What virtual place preference tests
In the rodent test, an animal chooses between two compartments after one of them has been paired with a reward.2 The test is usually run without the reward, and the classic result is more time in the compartment where the reward used to be.2 The preference is thought to reflect Pavlovian conditioning, in which a neutral place becomes preferred through its link with the reward.2 Linhardt and colleagues describe the test as a way to study approach and avoidance, the pull toward or away from places tied to good or bad outcomes.4
A virtual environment lets researchers build distinct places that people move through freely without leaving the laboratory.2 Astur and colleagues built their virtual version to carry this standard animal paradigm over to people, with real food as the reward.3
How the three versions run
Childs and colleagues tested 57 healthy men and women over two days, with money as the reward and the rooms shown on a computer screen.1 People see the rooms from a first-person view and move with the arrow keys and a mouse.1 Rooms A and B are the same size and hold similar furniture, a television, sofa, table and bookcase.1 They differ in the colors of the walls, carpet and sofa, the pictures on the walls, and the place and design of the furniture.1 A hallway leads into both rooms, and a doorway joins them.1
Practice the controls
Day 1 opens with a 60-second practice trial in a plain gray space, so that people learn the controls before they see either room.1
Pair one room with more money
People then complete six conditioning trials of 2 minutes each, shut in one room per trial, with 3 trials in each room.1 The order is counterbalanced, ABBAAB or BAABBA.1 They are told to move about and collect balloons, but money in fact arrives at random, signaled by a sound, whatever they do.1 Each payment is $0.15, and the payments add up to $4.35 per trial in the high reward room and $0.60 in the low reward room.1 Which room pays more is assigned at random for each person.1
Test the preference without money
Each preference test lasts 1 minute, with no money and no balloons.1 A test starts in the hallway, and once people enter a room they cannot return to the hallway, though they can cross between the rooms through the doorway.1 The program records the time spent in each room.1 People also rate how much they like a picture of each room, from 0 for dislike very much to 9 for like very much.1 A second scale under the two pictures runs from 5 for one room through 0 for neither to 5 for the other.1
Retest the next day
People return 24 hours later for four more 1-minute preference tests, which measure how well the preference is retained and how it extinguishes, or fades, without reward.1
Molet and colleagues used two houses on either side of a road, where each visit began.2 The houses differed in the layout, size and shape of their rooms, floor colors or wallpaper, furniture, lamps and carpets.2 A session had a 1-minute familiarization phase, one 2-minute visit to each house, and a test with no time limit.2 People moved in first person at a constant speed, with the arrow keys to move and the mouse to look around, on a computer screen.2 An observer blind to the conditions scored the time in each house from video recordings of the screen.2
Astur and colleagues ran 6 pairing sessions on day 1, 3 with real candy in one room and 3 with no food in the other.3 People returned the next day for a 5-minute test with free access to both rooms.3
What the test has shown in people
Music
In the first experiment of Molet and colleagues, one house came with consonant music and the other with static noise, 2 minutes each.2 Given a free choice, people spent 85.1% of their time in the music house, more than the 50% expected by chance.2 Thirteen of 15 people visited the music house first, and all 15 named it as their favorite.2 None of them thought the music had influenced their choice.2 In a second experiment, dissonant music replaced the consonant music, and people spent 95.7% of their time in the house with static noise.2 Molet and colleagues call this a conditioned place aversion.2 They conclude that virtual environments can extend place preference and aversion research to people.2
Food and hunger
In the food version of Astur and colleagues, participants who had been food restricted spent more time in the room paired with candy.3 They also picked it when asked which room they liked best.3 Participants who were not food restricted showed no place preference, in time spent or when asked.3 Astur and colleagues conclude that a place preference can be reliably set up in people, but that it depends on hunger.3
Eating attitudes
Astur and colleagues repeated the food version with 80 food-restricted undergraduates, who again preferred the room paired with candy.5 Participants also completed the Eating Attitudes Test, a standard screening questionnaire for eating disorder risk.5 The stronger a participant’s place preference, the higher their score on the test’s dieting subscale.5 Ratings of the room without food fell as dieting scores rose.5 The authors suggest that parts of eating disorder risk can influence how strongly people condition to places tied to food.5
Money
With money as the reward, people spent more time in the room that paid more, and they said they liked it more and preferred it.1 The preference was short-lived, and 24 hours later people no longer showed it in time spent or in their ratings.1 Results were similar at the two sites that ran the task.1
Across studies
Linhardt and colleagues pooled 17 human place preference studies published before June 2020, run in either virtual or real environments.4 Across them, time spent in the reward-paired place showed a medium effect size (g = 0.62), and verbal self-ratings a small one (g = 0.33).4 The pool mixes virtual and real environments, so these effects describe human place preference in general.4 Results varied substantially from study to study.4
How it compares with the rodent place preference test
The virtual test keeps the rodent design, two distinct places, a reward paired with one of them, and a choice test with the reward gone.2,1 Both versions are scored by the time spent in each place.2,1 Linhardt and colleagues see great potential for the human test to study approach and avoidance, in direct analogy to animal studies.4 Molet and colleagues add that the virtual procedure could be adapted to brain imaging such as MRI and PET, which need the head held still.2
Human versions add explicit measures, such as which room a person likes best.3,1 In the music study, people showed no built-in preference for either house, unlike the animals in most laboratory studies of place preference.2 Molet and colleagues used an unbiased design, with the music assigned to either house by counterbalancing, and note an emerging consensus that this design is preferable.2 Rodent tests typically last 15 minutes, while these human tests ran from 1 minute to no time limit.2,1 Molet and colleagues note that music alone does not appear to produce a place preference in rats, so humans may be unique in finding music rewarding.2 The money protocol had no pre-test before conditioning, which Childs and colleagues list as a limitation.1
Build a virtual place preference task with ConductScience VR
We write the task software in house and build each place preference task to the protocol you plan to run.
- The rooms, their colors, pictures and furniture, and the hallway and doorway between them
- Practice, conditioning and test trials, with your room order, trial lengths and which room carries the reward
- Rewards in the task, such as music in one room or a sound with each payment, and a prompt for the experimenter when a real reward such as candy is due
- Time in each room, room entries, paths and room ratings for every test, saved for analysis
- A desktop screen with a keyboard and mouse or a joystick, or a headset
- The software alone, or a package with the headset and computer
We also build teaching versions, so students in a class can test each other on the task.
Related pages
Rodent behavior tests rebuilt in virtual reality for people, and how we build them.
Our rodent place preference box, with two chambers whose walls and floors differ, for mice and rats.
Virtual context fear conditioning
A mild electric shock tied to one virtual room, measured by startle, ratings and avoidance.
Set up a rodent behavior testing room
How to keep light, sound, smells, handling and timing the same for every animal.
Questions
- Is virtual place preference the same test as the rodent one?
- It keeps the rodent design, with a reward paired to one place and a choice test given without the reward.2,1 Human versions also ask people which room they like best.3
- What rewards produce a virtual place preference?
- Consonant music, candy and money have each produced one.2,3,1 Candy worked only when participants had been food restricted.3
- How long does a virtual place preference last?
- It depends on the version. With candy, food-restricted people preferred the paired room at a test the next day.3 With money, the preference was short-lived, gone when people were retested 24 hours later.1
- Can the test measure aversion as well as preference?
- Yes, people spent less time in a house paired with dissonant music, a conditioned place aversion.2
References
- Childs E, Astur RS, de Wit H. Virtual reality conditioned place preference using monetary reward. Behavioural Brain Research. 2017. doi:10.1016/j.bbr.2017.01.019
- Molet M, Billiet G, Bardo MT. Conditioned place preference and aversion for music in a virtual reality environment. Behavioural Processes. 2013. doi:10.1016/j.beproc.2012.10.001
- Astur RS, Carew AW, Deaton BE. Conditioned place preferences in humans using virtual reality. Behavioural Brain Research. 2014. doi:10.1016/j.bbr.2014.03.018
- Linhardt M, Kiser DP, Pauli P, et al. Approach and avoidance beyond verbal measures: A quantitative meta-analysis of human conditioned place preference studies. Behavioural Brain Research. 2022. doi:10.1016/j.bbr.2022.113834
- Astur RS, Palmisano AN, Hudd EC, et al. Pavlovian conditioning to food reward as a function of eating disorder risk. Behavioural Brain Research. 2015. doi:10.1016/j.bbr.2015.05.016
More virtual reality tests
ConductScience VR
ConductScience VR
ConductScience VR
ConductScience VR
ConductScience VR
ConductScience VR
ConductScience VR
ConductScience VR
ConductScience VR
