visRecog {dbd}

Visual recognition data.

Description

Counts of successes in visual recognition memory for large and small binary pictures.

Usage

data("visRecog")

Format

A data frame with 30 observations on the following 4 variables.

deck

An integer vector indicating which of two decks of cards, bearing graphic images, was used in the given experiment.

subject

An integer vector indexing the (human) subjects in the experiments.

tot5

An integer vector whose entries are counts of successes when the cards used consist of a 5 \times 5 grid of “facets”.

tot10

An integer vector whose entries are counts of successes when the cards used consist of a 10 \times 10 grid of “facets”.

Details

Adult subjects were shown a series of cards, each bearing a simple graphic image. Each image resembled one face of a Rubik's cube, formed of either a 5x5 or a 10x10 grid of facets, each facet being either black or white. Later, each subject was shown a series of 20 similar cards, exactly 10 of which had been shown to the subject previously. The subject's task was to identify each image as a new one, or as a previously seen one. The response variable tot5 is the number of correct identifications, out of 20, for the 5 \times 5 cards. Similarly the variable tot10 is the number of correct identifications for the 10 \times 10 cards.

Subjects 21–30 were (deliberately) tested with a different set of cards than subjects 1–20, to ensure that results were not a function of the original deck of cards. (This seems to have no actual relevance.)

Source

The data are taken from the paper sited in References below. They were provided by a generous email correspondent who prefers to remain anonymous.

References

Green, D. M., and Purohit, A. K. (1976). Visual recognition memory for large and small binary pictures. Journal of Experimental Psycholology: Human Learning and Memory 2, pp. 32–37.

Examples

dbfit5  <- with(visRecog,mleDb(tot5,20,TRUE))
dbfit10 <- with(visRecog,mleDb(tot10,20,TRUE))
set.seed(42) # To get repeatable Monte Carlo p-values.
print(gof(dbfit5,obsd=visRecog[["tot5"]],MC=TRUE)$pval)   # 0.86
print(gof(dbfit10,obsd=visRecog[["tot10"]],MC=TRUE)$pval) # 0.68
bbfit5  <- with(visRecog,mleBb(tot5,20))
bbfit10 <- with(visRecog,mleBb(tot10,20))
set.seed(42) # To get repeatable Monte Carlo p-values.
print(gof(bbfit5,obsd=visRecog[["tot5"]],MC=TRUE)$pval)   # 0.94
print(gof(bbfit10,obsd=visRecog[["tot10"]],MC=TRUE)$pval) # 0.70