Ah, Halloween. 

Tonight's the one night of the year when children can don their favorite costumes, temporarily transforming themselves into little Olafs and Annas and Elsas, and giddily skip around their local neighborhoods, knocking on doors, cheerfully thrusting their buckets out towards those who answer, grinning their wide, toothy smiles, and screaming, "TRICK OR TREAT!"

If they're really lucky -- ohhhh, if they really hit the trick-or-treaters' jackpot -- they'll stumble upon the Holy Grail of trick-or-treating quests. They'll find a house -- maybe even a few -- where the owners just left a bowl of candy out for the taking, likely with a sign requesting, ever-so-sweetly, "Please Take Only One!"

Ha. As if that'll happen.

However, even those of us who are home tonight to hand out candy to the neighborhood kiddos know that the "Please take only one!" request doesn't...always get respected. All you have to do is turn your back or show a sign of weakness, and BAM. It can be a free-for-all around that candy bowl.

This trick-or-treating dilemma is such a familiar experience for so many people that in 1976, a team of psychologists in Seattle led by Ed Diener thought of this annual experience and decided to use it to conduct a little experiment on deindividuation.

Deindividuation, briefly, is a psychological theory stating that when we are in groups, we lose some of our self-awareness and become more likely to act in disinhibited, non-socially-normative ways. It's a little more complicated than "peer pressure" or "going with the flow." It's more about the fact that (a) when we're surrounded by people, it makes us feel more physically amped up, and (b) when we feel anonymous, we feel less self-conscious and less responsible for our actions. The combo of higher physiological arousal and lower self-awareness means that in these situations, we're more likely to lose our inhibitions and do some brazen things that we otherwise might not.

Going back to our story above, you can think of "taking more candy than you're supposed to take" as a Halloween-specific example of disinhibited, socially-inappropriate, transgressive behavior -- you're breaking the rules that you've been given, and you're doing something that you're not supposed to be doing. Diener and his colleagues wanted to see if they could play around with some of the different situational factors that are thought to influence the strength of deindividuation's effects, and see whether or not these factors actually had any influence on how many kids would steal some extra sweets if they got the opportunity.

If deindividuation gets enhanced when people feel less self-aware and more physiologically aroused, anything that (a) increases anonymity and/or (b) puts them in a group will do the trick -- and, conversely, anything that makes people feel more identifiable, individuated, or singled-out will dampen it. The presence (or absence) of groups was out of the researchers' control -- kids either showed up alone or in groups of friends. But what the researchers could do was play around with that other factor -- anonymity -- and see how it impacted the trick-or-treaters differently based on whether or not they were there with friends.

When the kids showed up at the doors of the 27 Seattle homes included in this study, there was always an experimenter there to open the door -- an experimenter who didn't live in the neighborhood, and thus was unfamiliar to the kids. For half of the 1,000+ trick-or-treaters that unknowingly participated in this study on that Halloween night in the late 1970s, the experimenter simply greeted them warmly, told them that they could each take exactly one candy from the bowl sitting in front of them in the hallway, and then said that she had to go back to another room in the house to finish up some work. For the other half, the experimenter did the same thing -- except that before she offered them the candy and made her exit to the back of the house, she explicitly asked all of the trick-or-treaters what their names were and where they lived. After repeating the information back to them -- making it clear that she knew exactly who they were and where they lived -- she told them about the "One Candy Only" rule and made her exit. Unbeknownst to the kids, though, her "work" really just took her back behind a little hidden peephole -- where she could see exactly what those lil' costumed mischief-makers were doing.

What do you think happened?

As the researchers had expected, the two factors of anonymity and group presence made a huge difference in how good -- or bad -- the trick-or-treaters ended up being. Only 7.5% of the trick-or-treaters who showed up to the houses alone and provided their names and addresses ended up stealing extra candy out of the bowl. On the opposite end of the spectrum, however, were the naughty little trick-or-treaters who showed up in groups and stayed anonymous. A whopping 57% of those trick-or-treaters ended up sneaking a few extra candies from the bowl when they thought nobody was looking.

"But wait," an astute reader might note right about now. "Are we sure this is specifically about deindividuation? Maybe it's really just about peer pressure?" 

Well, it is. A little. But also not.

As you can see above, the trick-or-treaters' behavior was shaped by what the first kid in the group ended up doing. If the "leader of the pack" who got first dibs at the candy bowl decided to palm a few extra sweets, the majority of her comrades ended up deciding to do the same. On the other hand, though, if the first kid decided to be good and only take one candy, a whopping 90% of the rest of the group followed suit. So, there's definitely evidence for modeling here -- the kids did watch what the first trick-or-treater did, and used his behavior as a guide. If he stole, they stole. If she didn't, they didn't.


...that can't possibly be the entire story, because that first kid to go up still had to make a decision about what to do, and that decision was made in the absence of information about other group members (since, well, he was the first kid to get a crack at the candy bowl). As you can see in the graphs above, when the researchers went back and looked specifically at what the first kid in a group decided to do, they found that significantly more of those trick-or-treaters than expected (based on the solo-kid rates) stole extra candies -- especially when they had the added bonus of being anonymous.

This clues us into the fact that there is something special about merely being in a group, even if you aren't strictly modeling your own behavior after the other group members -- and there is something special about being anonymous, especially when you're amped up from having a bunch of your buddies around you.

The moral of the story here? If you want your trick-or-treaters to obey the one-candy-only rule -- well, keeping a close eye on your candy bowl is probably your best bet. But you can also innocently ask your trick-or-treaters what their names are and where they live.

Research suggests that if you just ask these two simple questions, you will reap great rewards -- in the form of a gloriously non-depleted candy bowl come November 1st.


Diener, E., Fraser, S.C., Beaman, A.L., & Kelem, R.T. (1976). Effects of deindividuation variables on stealing among Halloween trick-or-treaters. Journal of Personality and Social Psychology, 33, 178-183.

Image Credits

Candy Corn by Evan Amos via Wikimedia Commons

Trick or Treaters by Belinda Hankins Miller via Wikimedia Commons