Lewis O. Harvey, Jr.

Amendment I: Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech, or of the press; or the right of the people peaceably to assemble, and to petition the government for a redress of grievances. (click here). Humata, Hukhta, Huvarshta.

My Software

Revised on Monday 30 November, 2015 9:44 Mountain Time

Click on the highlighted links to download the appropriate software. I am using Metrowerks CodeWarrior Pro 9 to produce both Macintosh and Windows 95/98/NT programs from the same ANSI C/C++ source code. Each file set is in the form of a zip archive that is automatically unpacked by both Mac OS X and Windows operating systems.

  • RscorePlus: Version just for the Macintosh, written in objective-C and C++ using the Macintosh Cocoa interface. It features a convenient modern graphical user interface. It will run under Mac OS 10.5 (Leopard) through 10.10 (Yosemite). At the moment it is continuously under development. It includes four additional models that can be fit (high threshold, low threshold, low-high threshold and dual process models). It uses the same input text file as the older version so you can use your old RscorePlus input files unchanged. The graphic displays on the interface are fillers for the moment. They are produced by core-plot and I anticipate that I will make them functional in my free time. Give it a try. The documentation below is for the Cocoa version and is also included in the zip archive.

    Current Mac Cocoa version: v 5.8.3, 13 October 2013

    Download RscorePlus_Mac_Cocoa.zip (for Intel Macs running OS 10.5 and later)
    Download RscorePlus_doc_5_8_3.pdf (Adobe pdf file)

    RscorePlus: A Macintosh/Windows classic program that fits a multiple-distribution, variable-criterion signal detection theory model to confidence-rating data. The classic version will fit Gaussian, Logistiic, Extreme Minimum and Extreme Maximum probability distributions to data along with the (mostly) discredited High Threshold Model for laughs. It was originally derived from Dorfman & Alf's (1969) RSCORE program. The package contains the classic program, some sample data files, an R script file for graphing, and a documentation file in Adobe pdf format. Double-click on the link below to download the documentation file by itself. You can get the free Adobe Reader, if you need it, from Adobe or use other pdf readers (e.g., Apple's Preview that comes withMac OS X). The Mac version requires Rosetta and therefore will not run under Lion and Mountain Lion (see new, improved version above)

    Current classic version: v 5.6.1, 22 September 2010.

    Download RscorePlus_Mac_classic.zip (for Macintosh with Rosetta installed)
    Download RscorePlus_Win.classic.zip (for Windows)
    Download RscorePlus_doc_5_6_1.pdf (Adobe pdf file)

  • PsychoFit: A Macintosh/Windows program that fits s-shaped psychometric functions to psychophysical data using a maximum-likelihood technique. The psychometric functions available are logistic, Weibull, Gaussian integral, cumulative Poisson and step. The program is also included in the above MLPest package. The documentation is pretty weak, and the interface for the Mac carbon and Windows versions is unix-like command line. PsychoFit can accept date in one of three input formats: 1) number of correct and number of incorrect for each stimulus; 2) number of correct and total number of trials for each stimulus; and 3) the probability correct and the total number of trials for each stimulus; A sample data file with the same data in each of the three formats, is included, so copy its format. PsychoFit produces, among others, files with graphic information suitable for importing in a graphic program for plotting. A sample R script is included that makes psychometric function plots of the observed data, the smooth predicted psychometric function of the best-fitting model, and marker lines. I am preparing a users' manual that will provide better documentation. The current Macintosh files include a carbon version that will run under earlier versions of Mac OS X, as well as on earlier versions of the Mac OS using the CarbonLib.There is also a version using a Cocoa GUI that runs on Lion, Mountain Lion, and Mavericks. I'll try to make the Mac Cocoa version more elegant when I have the spare time (whenever that might be!). The Windows version should run on Windows 95 and higher with no problems. The computational algorithms use the same code in all three versions.

    Current version of PsychoFit: 20 June 2014

    Download PsychoFit_Mac_Carbon.zip (for Macintosh - 140 KB)
    Download PychoFit_Mac_Cocoa.zip (for Macintosh using GUI - 292 KB)
    Download PsychoFit_Win.zip (for Windows --- 322 KB)

  • MLPest: Maximum-Likelihood Parameter Estimation by Sequential Testing
    A package of subroutines and classes written in ANSI C and C++ to implement the maximum-likelihood adaptive staircase method of estimating psychophysical thresholds as described in Harvey (1986, 1997). The package includes all source code and projects for CodeWarrior Pro 9. The source code itself should compile and run on any computer with an ANSI complient C/C++ compiler. Several demonstration programs illustrating the use of the routines and classes are included in the package.

    Current version of MLPest: 18 May 2004.

    Download MLPest.zip (2.4 MB)

  • CSF: A Macintosh program that measures a contrast sensitivity function using the MLPest technique. It requires subroutines from Denis Pelli's VideoToolbox and the Pelli video attenuator connected to the monitor. The CSF on my home page was measured with this program, requiring about 23 2AFC trials per spatial frequency.
    Download CSF

  • Sensory Tester: A Macintosh program for measuring taste and smell sensory thresholds (or any thresholds where a limited number of stimuli must be presented manually). The trial-by-trial results and the likelihood functions are displayed graphically.
    Download Sensory Tester

Valid XHTML 1.0 Transitional

Created with Dreamweaver CS6 on a MacBook Pro