Editing Cognition (section)

=== Common types of tests on human cognition ===
====Serial position====
The ''[[Serial-position effect]]'' is meant to test a theory of memory that states that when information is given in a serial manner, we tend to remember information at the beginning of the sequence, called the ''primacy effect'', and information at the end of the sequence, called the ''recency effect''. Consequently, information given in the middle of the sequence is typically forgotten, or not recalled as easily. This study predicts that the recency effect is stronger than the primacy effect, because the information that is most recently learned is still in working memory when asked to be recalled. Information that is learned first still has to go through a retrieval process. This experiment focuses on human memory processes.<ref>{{cite journal | vauthors = Surprenant AM | title = Distinctiveness and serial position effects in tonal sequences | journal = Perception & Psychophysics | volume = 63 | issue = 4 | pages = 737–745 | date = May 2001 | pmid = 11436742 | doi = 10.3758/BF03194434 | doi-access = free }}</ref>

====Word superiority====
The ''[[word superiority effect]] experiment'' presents a subject with a word, or a letter by itself, for a brief period of time, i.e. 40&nbsp;ms, and they are then asked to recall the letter that was in a particular location in the word. In theory, the subject should be better able to correctly recall the letter when it was presented in a word than when it was presented in isolation. This experiment focuses on human speech and language.<ref>{{cite journal | vauthors = Krueger LE | title = The word-superiority effect and phonological recoding | journal = Memory & Cognition | volume = 20 | issue = 6 | pages = 685–694 | date = November 1992 | pmid = 1435271 | doi = 10.3758/BF03202718 | doi-access = free }}</ref>

====Brown–Peterson====
In the ''[[Brown–Peterson cohomology]] experiment'', participants are briefly presented with a [[trigram]] and in one particular version of the experiment, they are then given a distractor task, asking them to identify whether a sequence of words is in fact words, or non-words (due to being misspelled, etc.). After the distractor task, they are asked to recall the trigram from before the distractor task. In theory, the longer the distractor task, the harder it will be for participants to correctly recall the trigram. This experiment focuses on human [[short-term memory]].<ref>{{cite journal |vauthors=Nairne J, Whiteman H, Kelley M |title=Short-term forgetting of order under conditions of reduced interference |journal=Quarterly Journal of Experimental Psychology A |year=1999 |volume=52 |pages=241–251 |doi=10.1080/713755806 |s2cid=15713857 |url=http://www1.psych.purdue.edu/~nairne/pdfs/32.pdf |access-date=2018-01-09 |archive-date=2022-07-30 |archive-url=https://web.archive.org/web/20220730173204/http://www1.psych.purdue.edu/~nairne/pdfs/32.pdf |url-status=dead }}</ref>

====Memory span====
During the ''[[memory span]] experiment'', each subject is presented with a sequence of [[Stimulus (physiology)|stimuli]] of the same kind; words depicting objects, numbers, letters that sound similar, and letters that sound dissimilar. After being presented with the stimuli, the subject is asked to recall the sequence of stimuli that they were given in the exact order in which it was given. In one particular version of the experiment, if the subject recalled a list correctly, the list length was increased by one for that type of material, and vice versa if it was recalled incorrectly. The theory is that people have a memory span of about seven items for numbers, the same for letters that sound dissimilar and short words. The memory span is projected to be shorter with letters that sound similar and with longer words.<ref>{{cite journal | vauthors = May CP, Hasher L, Kane MJ | title = The role of interference in memory span | journal = Memory & Cognition | volume = 27 | issue = 5 | pages = 759–767 | date = September 1999 | pmid = 10540805 | doi = 10.3758/BF03198529 | doi-access = free }}</ref>

====Visual search====
In one version of the ''[[visual search]] experiment'', a participant is presented with a window that displays circles and squares scattered across it. The participant is to identify whether there is a green circle on the window. In the ''featured'' search, the subject is presented with several trial windows that have blue squares or circles and one green circle or no green circle in it at all. In the ''[[Conjunctive tasks|conjunctive]]'' search, the subject is presented with trial windows that have blue circles or green squares and a present or absent green circle whose presence the participant is asked to identify. What is expected is that in the feature searches, reaction time, that is the time it takes for a participant to identify whether a green circle is present or not, should not change as the number of distractors increases. Conjunctive searches where the target is absent should have a longer reaction time than the conjunctive searches where the target is present. The theory is that in feature searches, it is easy to spot the target, or if it is absent, because of the difference in color between the target and the distractors. In conjunctive searches where the target is absent, reaction time increases because the subject has to look at each shape to determine whether it is the target or not because some of the distractors if not all of them, are the same color as the target stimuli. Conjunctive searches where the target is present take less time because if the target is found, the search between each shape stops.<ref>{{cite journal| vauthors = Wolfe J, Cave K, Franzel S |title=Guided search: An alternative to the feature integration model for visual search|journal=Journal of Experimental Psychology: Human Perception and Performance|year=1989|volume=15|pages=419–433|doi=10.1037/0096-1523.15.3.419 |issue=3 |pmid=2527952|citeseerx=10.1.1.551.1667 }}</ref>

====Knowledge representation====
The [[semantic network]] of [[Knowledge representation and reasoning|knowledge representation]] systems have been studied in various paradigms. One of the oldest paradigms is the [[leveling and sharpening]] of stories as they are repeated from memory studied by [[Frederic Bartlett|Bartlett]]. The [[semantic differential]] used [[factor analysis]] to determine the main meanings of words, finding that the [[ethical value]] of words is the first factor. More controlled experiments examine the categorical relationships of words in [[free recall]]. The hierarchical structure of words has been explicitly mapped in [[George Armitage Miller|George Miller]]'s [[WordNet]]. More dynamic models of semantic networks have been created and tested with computational systems such as [[neural network (machine learning)|neural networks]], [[latent semantic analysis]] (LSA), [[Bayesian analysis]], and multidimensional factor analysis. The meanings of words are studied by all the disciplines of [[cognitive science]].<ref>{{cite journal | vauthors = Pinker S, Bloom P | title = Natural language and natural selection. |date=December 1990 |journal=Behavioral and Brain Sciences |volume=13 |issue=4 |pages=707–727 |doi=10.1017/S0140525X00081061 | s2cid = 6167614 }}</ref>