Easy roommate behaviors to change through operant conditioning?

September 30, 2009 6:09 PM

What are some examples of how I could change my roommate's behavior through operant conditioning?

This is just for a little experiment for my psychology class, so I don't already have in mind anything specific I want change about my roommate. Although we have not become very close friends, we are on quite good terms and I like her! So I'm just looking for random ideas about how I might use positive or negative reinforcement to change some mundane behavior of hers. Do you have any ideas of what behaviors would be easy to target?

posted by datarose to Grab Bag (19 answers total)

Are you serious? You want to conduct psychological experiments on your roommate? One that you like, even? You may want to rethink this.
posted by Cat Pie Hurts at 6:11 PM on September 30, 2009 [3 favorites]

It's probably ethical if datarose has informed consent from her roommate, but in an academic environment, it probably needs institutional review board approval before she can proceed. You need to talk to your professor about getting IRB approval before you undertake this.
posted by mr_roboto at 6:16 PM on September 30, 2009

As long as you stick to mundane behaviors, then I think this shouldn't be that bad, ethically speaking. You may want to consult with your professor, though.

Does your roommate have a laptop? Does she leave it on or off? Does she wear her shoes in the room? Does she take them off? Does she put them in a certain place when she takes off her shoes? Does she leave clothes strewn around the room? Does she text a lot on her phone? Does she listen to music on speakers? On headphones? These are all mundane behaviors that you could probably influence and not actually cause any long term psychological damage.
posted by baxter_ilion at 6:17 PM on September 30, 2009

If I lived with you as a roommate, was not a close friend of yours, but I liked you, and then I found out that you were performing a psychological experiment on me in order to see if you could get me to wear socks fewer days out of the week or eat more grilled cheese sandwiches, I would either ask you to move out or move out myself, and tell all of my friends the story of that complete weirdo I lived with for the the rest of my life.
posted by pazazygeek at 6:23 PM on September 30, 2009 [12 favorites]

Response by poster: Thanks baxter_ilion for a few suggestions so far!

I wasn't really expecting such a harsh reaction from the others. This is just a casual sort of assignment I got in my freshman-level introductory psychology class; it's not meant to be a serious experiment. It's only going to result in two weeks of me trying something and explaining what if anything, happened in a 3-page paper. Of course, my T.A. and professor will look over what I'm planning to do (and what everyone else in the class will be doing) before giving the go-ahead.
posted by datarose at 6:33 PM on September 30, 2009

I would react negatively to being someone's school project, especially unknowingly. I would highly advise reconsidering this.
posted by glip at 6:33 PM on September 30, 2009 [1 favorite]

You probably share a fridge. Condition her to keep her shit on the left.
posted by turgid dahlia at 6:41 PM on September 30, 2009

If this is an actual assignment, your teacher could be in deep trouble. Most institutions would not allow this (in the U.S., anyway).
posted by Wordwoman at 6:49 PM on September 30, 2009

I agree that this is a questionable idea, so maybe you could ask roommate's permission of a sort?

"Hey, so, for a psych class I have to test something on you. No, no! Nothing big. Just spending 2 weeks seeing if I can have an affect on your behavior. Like getting you to wear red socks more often without actually asking. I swear it won't be anything messed up but I can't say what it is. Is that cool?"

If my roommate asked me that, I would be a little paranoid but okay. It's not like it's a rigorous scientific experiment.

I also thought of the office clip, by the way.
posted by Solon and Thanks at 6:52 PM on September 30, 2009

You should condition your professor to favor one side of the room while teaching. If you get all of your classmates in on it, all they need to do is to be extra-attentive when prof is on one side of the room, and to show signs of disinterest when the prof is on the other side.

I've heard anecdotal tales of this working, but I've always wondered how effective it would be.

Of course, if the whole class does this, you would need to remind them to be subtle as the prof moves around, and to wait a few beats before responding to a change in location.

Of course, you may have to practice "deception" by getting your TA in on it, and writing it up as a room-mate behavior modification scheme just for the proposal so your prof doesn't find out.
posted by Maxwell_Smart at 6:55 PM on September 30, 2009 [3 favorites]

Operant conditioning is tough to do ethically since you do kind of need to do it without their awareness, but the less manipulative, the better. Why not try to duplicate Pavlov's lab-- get a candy dispenser that makes a noise when it dispenses, then only eat candy in front of her for a while. After a couple of days, maybe you'll be able to trigger the dispenser noise while she's out of sight (but within hearing range) and she'll come around pretty soon to get some candy because she's not-so-mysteriously got a hankering for it. If you have a fridge that's mostly used for drinks, you could put a bell on the door for the same effect.

For the sake of the paper, you probably just need a sound methodology. A subtle treatment with minor effects is probably good enough and won't creep out your roommate.
posted by parkerjackson at 7:00 PM on September 30, 2009

You should condition your professor to favor one side of the room while teaching.

This is a classic example, but seems more unethical than anything the OP might do since it might embarrass the prof in front of a whole class. I'd be tempted to flunk anyone who did it to me...
posted by parkerjackson at 7:02 PM on September 30, 2009

Sutherland's What Shamu Taught Me About Life, Love, and Marriage might give you some ideas. Here's a link to an article about the book. You can preview it here.
posted by Houstonian at 7:05 PM on September 30, 2009

Response by poster: To those of you who mentioned the clip from The Office: our professor actually used that clip in class to illustrate classical conditioning!

He also told us about the experiment of conditioning the lecturer to favor one side of the room for teaching (and our textbook mentioned/playfully suggested it). So that wouldn't work on this professor anymore, but it is a very good example.

Thanks for the responses s far. Again, this is only meant to be a light-hearted, short-term experiment, and it is an assignment I am required to do. I am aware of the possible ethical concerns, but it's not going to be anything overly creepy or at all serious. I'm just trying to brainstorm and seek out good ideas--ideas which would be ethical.
posted by datarose at 7:23 PM on September 30, 2009

unplug her coffee maker every day for a week, then the next week don't touch it. maybe she will habitually check the outlet every morning regardless.

I agree with Solon and Thanks; warn her first, maybe turn it into a friendly game of who can condition whom.
posted by Think_Long at 7:40 PM on September 30, 2009

Possibly this is actually the professor's project, to determine whether an entire class of freshman psychology students will be willing to perform psychological experiments on their roommates without inquiring into the ethical aspects of the assignment.
posted by palliser at 7:44 PM on September 30, 2009 [7 favorites]

If I was your roommate, and it truly was a mundane behavior change I would laugh when I was told. Getting your roommie to put her food on one side of the fridge or to put her toothbrush on the other side of the sink is not a big deal. This type of thing happens all the time accidentally. I had a habit from home of putting the silverware into the dishwasher business side down. My roommate did it the other way. We never spoke about it, but after I repeatedly would change it before running it to my way, he eventually went the same way. When I finally mentioned it to him to thank him, he didn't even realize he had done it. No big deal. I learned from him not to drink the rest of a coke from a can. He dipped. One sip out of a half full can on the table that turns out to be tobacco spit cures that inclination.
posted by JohnnyGunn at 8:38 PM on September 30, 2009

Perhaps this is something for MetaTalk, but most of these actions (unplugging a coffee maker every day for a week, then not; making a professor favor one side of the room when teaching), if done outside of a psychology class context would be called "pranks" and would not be subject to such ethical handwringing. Why does this inspire such ethical concern when done as part of a class? Just wondering...

(And yes, I am familiar with Milgram and the Stanford Prison Experiment)
posted by baxter_ilion at 8:41 PM on September 30, 2009

Baxter, I think everyone else is saying that it's generally not a good idea to condition your freshman psych students to the idea that they can experiment on other humans without IRB oversight.

It does seem problematic, as you can't really obtain consent to condition someone without biasing your results-- "Hey, can I train you to only wear red socks every other day?" is going to result in purple socks, orange socks, and God only knows what else in the end.
posted by fairytale of los angeles at 10:26 PM on September 30, 2009