Yet Another Cognitive Flaw (YACF)

[theVOID]

As part of my ongoing effort to become as rational as possible I have spent a great deal of time coming to terms with the various cognitive flaws that taint our perception of reality, these post are intended to share some of the more interesting ones with the members of the forum

The attribution of Attitudes

This one is striking to me as it is a cognitive bias that takes place even when we are directly informed of the bias, and fully aware of the situation in which it is taking place, first revealed in a study by Edward Jones and Victor Harris in 1967.

The setting of the experiment was as follows:

Participants are audience members invited to hear speeches on controversial issues, in this particular case on Abortion Rights and Fidel Castro, they are told in advance that the people giving the speeches for and against the various propositions are assigned the speech randomly and that one cannot rationally come to a conclusion about the attitudes of the person regarding the issue on which the spoke based on the position argued for in their speech.

Despite this people were unable to disassociate the attitudes presented in speeches from those of the person who spoke about them, in social situations after the speeches had taken place people who were anit-abortion displayed noticeable strong negative attitudes towards the person who spoke for abortion rights even on issues unrelated to the speech, same for people who are pro-choice displaying negative attitudes towards the person who argued against abortion rights and the various sides of the Castro Debate. There was also some effect regarding positive attitudes for which the person agreed, but these effects were much less prominent.

So, if a person reads a speech for with which you disagree, especially on a particular polarizing issue, you are likely to harbor strong negative attitudes towards them in other areas even when you have been informed ahead of time that they have been assigned this position at random.

Human Rationality.... FAIL.

Considering these attitude attributions take place, and affect our attitudes towards people on other issues even when we have no rational reason to believe they support this position, how likely are we to have a strong negative bias towards people on issue x and y because of their attitude on issue z?

I will look into this more but already I get a sense that this rationality fail could be responsible for the absolute failure of political discourse in resolving some issues for which the people actually agree much more closely than they perceive, because of an issue that taints their perception of each other that is in it's self not directly related to the other ones being discussed.

http://www.radford.edu/~jaspelme/443/spr...s_1967.pdf

[Anomalocaris]

How do you know the supposed bias in perception is not based in part of observation of rather subtle correlation? A so called bias needs only to have slightly better than random chance of being right for it to possible to possess net positive practical to it's holders.

[theVOID]

How do you know the supposed bias in perception is not based in part of observation of rather subtle correlation?

As in the audience noticed some auditory or visual queues on the part of the speakers?

If the audience picked up on queues they would have only shown these hostilities towards people who believed the argument in the speech they were given, there was no noticeable difference in how people were treated afterwards between people who's speeches contained arguments they agreed with compared to those that did not - This strongly suggests that it was a non-factor, if it was you would expect people would be able to tell who did and did not believe the arguments they gave, at least to a statistically significant extent, there was no such phenomenon presented.

There were other controls too, professional speakers, speakers giving either speech in front of different audiences etc, plus this study has been done numerous times since it's first conception, always with confirmatory results.

A so called bias needs only to have slightly better than random chance of being right for it to possible to possess net positive practical to it's holders.

"So called bias"? It's a fairly obvious bias, a subset of correspondence bias.

The study says nothing about whether or not it is generally useful to attribute these attitudes, which it is, in much the same way our predator detection is generally useful in an evolutionary sense, the study was about human rationality, a question of: Given that we are informed ahead of time that the attitudes are assigned at random (which is not the case in a the environment where we evolved these traits) can we effectively take into account that piece of information when coming to conclusion? The answer was a resounding no.

[Zenith]

---

As part of my ongoing effort to become as rational as possible I have spent a great deal of time coming to terms with the various cognitive flaws that taint our perception of reality, these post are intended to share some of the more interesting ones with the members of the forum

The attribution of Attitudes

This one is striking to me as it is a cognitive bias that takes place even when we are directly informed of the bias, and fully aware of the situation in which it is taking place, first revealed in a study by Edward Jones and Victor Harris in 1967.

The setting of the experiment was as follows:

Participants are audience members invited to hear speeches on controversial issues, in this particular case on Abortion Rights and Fidel Castro, they are told in advance that the people giving the speeches for and against the various propositions are assigned the speech randomly and that one cannot rationally come to a conclusion about the attitudes of the person regarding the issue on which the spoke based on the position argued for in their speech.

Despite this people were unable to disassociate the attitudes presented in speeches from those of the person who spoke about them, in social situations after the speeches had taken place people who were anit-abortion displayed noticeable strong negative attitudes towards the person who spoke for abortion rights even on issues unrelated to the speech, same for people who are pro-choice displaying negative attitudes towards the person who argued against abortion rights and the various sides of the Castro Debate. There was also some effect regarding positive attitudes for which the person agreed, but these effects were much less prominent.

So, if a person reads a speech for with which you disagree, especially on a particular polarizing issue, you are likely to harbor strong negative attitudes towards them in other areas even when you have been informed ahead of time that they have been assigned this position at random.

Human Rationality.... FAIL.

Considering these attitude attributions take place, and affect our attitudes towards people on other issues even when we have no rational reason to believe they support this position, how likely are we to have a strong negative bias towards people on issue x and y because of their attitude on issue z?

I will look into this more but already I get a sense that this rationality fail could be responsible for the absolute failure of political discourse in resolving some issues for which the people actually agree much more closely than they perceive, because of an issue that taints their perception of each other that is in it's self not directly related to the other ones being discussed.

http://www.radford.edu/~jaspelme/443/spr...s_1967.pdf

---

Yeah, all people tend to like the ones that agree with them and dislike (or even, hate) those that disagree with them.

Very important thing, and it happens everywhere: if you want to be a friend to somebody, you have to agree with EVERYTHING he does and believes (e.g. if he beats his wife daily, to say that women deserve being treated like this, that you'd do the same if you were him, etc.) - if it's "religion", then you should either say that what he believes is "true" and convert, or say that it is possible, but you don't like talking about religions, and seek other activity or discussion. On the other side, if you want to be an enemy for somebody, one of the best way to do it is to ACCUSE him of doing evil (e.g. she's your wife! You must love her and protect her, not beat her! You do something evil!) - if it's about religion, saying "your religion is crap!" makes you being hated.

Point is, it seems we are more driven by feelings than by reason.