Fig 1.
(A) Rorschach Blot One (with dimension D = 1.11), (B) a computer-generated fractal blot (D = 1.15).
Fig 2.
(A) A zoom-in on Blot Seven highlighting the extracted boundary with a black line. The scale bar corresponds to 1.6mm (19 pixels in the scanned image). (B) Rorschach Blot Five of width 17.5cm. (C) A schematic representation of the box-counting technique applied to Blot Five’s boundary. The box size shown represents the largest box size analyzed.
Fig 3.
Top three traces: Box-counting analyses applied to Rorschach Blot Five (black), Rorschach Blot Ten (red), and a computer-generated fractal boundary (blue).
Bottom three traces: the derivatives of the three scaling plots (see text for details).
Fig 4.
Computer-generated fractal boundaries.
The black pattern has a D value of 1.26 that closely matches that of Rorschach Blot Five, while the grey pattern has a much higher D value of 1.77. The black and white squares indicate the smallest and largest box sizes used in the fractal analysis of these boundaries.
Fig 5.
(A) Rorschach Blot Ten, (B) the boundaries of the blot edges.
Fig 6.
Number of percepts, n, induced by the five black blots plotted as a function of the D value of their boundaries.
The inset tabulates the D values of the five blots along with the corresponding data from Hertz [5] and Wirt and McReynolds [21] respectively. For the purpose of direct comparison, the raw n scores have been standardized as z-scores and are shown in parentheses. The horizontal dotted line represents the mean score for each of the two sets of data. The red and blue lines are linear fits to the respective data and the dashed lines represent the boundaries of 95% confidence intervals. Note that three of the blue data points (for Blots 1, 5 and 6) are obscured by the equivalent red points.
Fig 7.
The standardized number of reported percepts, n, induced by the computer-generated fractal patterns plotted as a function of their D value.
One set of computer-generated images is illustrated above the graph. The dotted lines represent the boundaries of 95% confidence intervals. The inset tabulates the D values of the images along with the corresponding average number of reported perceived shapes.
Fig 8.
(A) Rorschach Blot Seven, (B) Rorschach Blot Seven with the fractal features removed.
Fig 9.
Koch curves quantified by D = 1.26.
The exact fractal (top trace) is morphed into the statistical fractal (bottom trace). See the text for details.