Knoblauch Kenneth - Modeling Psychophysical Data in R

(1)

(2)

In this chapter we illustrate how R can be used to explore and analyze psychophysical data. We examine a data set from the classic article by Hecht et al. [] in order to introduce basic data structures and functions that permit us to examine and model data in R .

The best way to read this book is sitting at a computer with R running, typing in the examples as they occur. If you do not already have R installed on your computer, see Appendix A for instructions on how to install it. If you have properly installed R and launched it by double clicking on an icon or by running the command R from a terminal command line, you should see some initial information printed out about the version followed by some initial suggestions of things to explore, demo() , help() , etc. This is followed by > on a line by itself. This is the command line prompt and indicates where you type in commands. If you enter a new line (carriage return) before the end of a command, the prompt will become a + , indicating a continuation of the previous line. You will continue to receive continuation lines until the command is syntactically complete, at which point R will attempt to interpret it. Comments in R are preceded by a # . Everything on the line to the right of a comment character is ignored by R . Lets proceed with the example.

In a classic experiment, Hecht et al. [] estimated the minimum quantity of light that can be reliably detected by a human observer. In their experiment, the observer, placed in a dark room, was exposed at regular intervals to a dim flash of light of variable intensity. After each flash presentation, the observer reported seen or unseen. The data were summarized as the percentage of reports seen for each intensity level used in the experiment, for each observer.

Before collecting these data, Hecht et al. performed extensive preliminary analyses in order to select viewing conditions that are described in their article, for which human observers are most sensitive (requiring the fewest number of quanta to detect) and a detailed analysis of their choices is presented in the opening chapter of Cornsweets text [].