Visual recognition capability does not match with IQ

bylondinrikard Fröberg

Ability to deal with cognition has been thought to be related to intelligence so far, Professor Isabel Gauthier of Vanderbilt University psychology department and others etc. devised and carried out a test to measure the visual recognition ability, and the result was I found that it is not consistent with IQ.

General object recognition is specific: Evidence from novel and familiar objects - ScienceDirect
http://www.sciencedirect.com/science/article/pii/S0010027717301361

Visual intelligence is not the same as IQ | Research News @ Vanderbilt | Vanderbilt University
https://news.vanderbilt.edu/2017/11/07/visual-intelligence-iq/


If it is a specific type of object already known, such as "face" "car" "birds", we know that the discrimination ability will rise as the learning time gets longer, but if it is a completely unknown object I can not predict how much it will be able to be identified.

Professors Gauthier and colleagues developed "Nobelm Object Memory Test (NOMT)" using unknown objects to visually measure the ability to identify objects. Unknown objects to be identified are three kinds of objects called "greebles", "ziggerins", "sheinbugs" generated by computers, and the subjects are 100 "amateurs" who collected using Amazon crowdsourcing.

This is "greebles" "sheinbugs" "ziggerins". Three types are listed here as an example here, but the subjects learn a total of 6 items, 3 types × 2 subjects.

by Isabel Gauthier / Vanderbilt

In the test body, three objects, one for each object learned by the subject and two for the unknown appearance, were set as a set, and the subject was chosen from among the three subjects.

As a result of performing the discrimination test with more than 2000 objects, the discrimination result in one kind of "greebles", "sheinbugs", "ziggerins" becomes similar to the other two kinds of discrimination results, "different unknown object It is predictable how much it will be able to distinguish each even if there is. " This result shows that NOMT functions as a test to measure visual ability when learning unknown categories.

In addition, IQ related tests were also conducted at the same time, and I also found that there is no relationship between IQ and the visual recognition ability measured by NOMT. Professor Gauthier was also surprised at the result because cognitive ability has been considered in connection with intelligence.

Professor Gauthier says that NOMT is planning to find out how it can be applied in the real world as visual perception ability is required not only in work but also in various places.