M H Gräf1. 1. Department of Strabismology and Neuroophthalmology, University of Giessen, Germany.
Abstract
BACKGROUND: If visual acuity (VA) and/or visual field (VF) is claimed to be worse than in reality, it can be difficult to estimate the actual VA/VF. This paper describes a method of proving malingering and of estimating the actual VA by statistical evaluation of the subjective responses to discrete visual stimuli. METHODS: VA is measured by using a Landolt or Snellen optotype with four possible directions. There are n (> or = 16) questions at each acuity level. Each direction of the optotype occurs with the same frequency. The sequence is not predictable. In testing the VF, the stimulus is presented in an unpredictable sequence but with the same frequency at each of four distinct locations outside the claimed field at the horizontal and the vertical meridian. The field is divided into four quadrants defined by the 45 degrees-225 degrees and 135 degrees-315 degrees meridians. The candidate is requested to search for the light. The target quadrant of the first eye movement is recorded. At each VA level, the distribution function of the binomial formula allows estimation of the probability Pc of < or = kc and > or = kc correct answers to n questions by mere accident. Regarding the VF, the direction of the first saccade at each stimulus presentation can be processed accordingly. RESULTS: If n = 32 and kc = 0,1,2, the chance P that this pattern is merely accidental is about 0.0001, 0.001, 0.007. Values of P < or = 0.01 strongly suggest that the answers were voluntarily wrong. Preference for the opposite direction can also point to psychogenic influence: If ko = 15,16,17,18, P is about 0.006, 0.002, 0.0006, 0.0002. CONCLUSION: Systematically false statements can yield valuable information about the actual visual functions.
BACKGROUND: If visual acuity (VA) and/or visual field (VF) is claimed to be worse than in reality, it can be difficult to estimate the actual VA/VF. This paper describes a method of proving malingering and of estimating the actual VA by statistical evaluation of the subjective responses to discrete visual stimuli. METHODS: VA is measured by using a Landolt or Snellen optotype with four possible directions. There are n (> or = 16) questions at each acuity level. Each direction of the optotype occurs with the same frequency. The sequence is not predictable. In testing the VF, the stimulus is presented in an unpredictable sequence but with the same frequency at each of four distinct locations outside the claimed field at the horizontal and the vertical meridian. The field is divided into four quadrants defined by the 45 degrees-225 degrees and 135 degrees-315 degrees meridians. The candidate is requested to search for the light. The target quadrant of the first eye movement is recorded. At each VA level, the distribution function of the binomial formula allows estimation of the probability Pc of < or = kc and > or = kc correct answers to n questions by mere accident. Regarding the VF, the direction of the first saccade at each stimulus presentation can be processed accordingly. RESULTS: If n = 32 and kc = 0,1,2, the chance P that this pattern is merely accidental is about 0.0001, 0.001, 0.007. Values of P < or = 0.01 strongly suggest that the answers were voluntarily wrong. Preference for the opposite direction can also point to psychogenic influence: If ko = 15,16,17,18, P is about 0.006, 0.002, 0.0006, 0.0002. CONCLUSION: Systematically false statements can yield valuable information about the actual visual functions.