![](bg2.png)
procedures, such as comprehension and efficiency in
reading text, which considers very simplistic quanti-
tative metrics like the count of words read per minute
(Irlen, 1990).
There are still few studies that identify the dis-
criminating characteristics present in the ocular signs
of people with the syndrome (Romera et al., 2019)
(Guimar
˜
aes et al., 2020) (de Faria, 2011). However,
recent approaches that explore eye-tracking tools are
presenting new findings on the behavior of individuals
with the syndrome, making it possible to identify pat-
terns from data such as eye movements, regressions,
saccades and pupil diameter changes (Romera et al.,
2019).
2.2 Mental Effort and Pupillary
Measurement
The human eye is composed of several parts that, by
means of visible light, capture the visual informa-
tion present in the environment. One of these parts
is the pupil, which is an orifice through which light
enters and which has a diameter that can be regulated
by two iris muscles: the sphincter muscle, which is
innervated by the parasympathetic part of the ANS
(Autonomic Nervous System) and is responsible for
pupil constriction, and the dilator muscle, which is
innervated by the sympathetic part of the ANS and
is responsible for pupil dilation (Marieb and Hoehn,
2007). The constriction and dilation are caused by
two factors: (1) to control the luminosity that enters
the eye, protecting the photoreceptor cells present in
the retina, which are responsible for sending informa-
tion by means of electrical impulses to the brain via
the optic nerves and, (2) by an involuntary reflex of
the ANS during the transition between states of at-
tention and rest that regulate the amount of informa-
tion captured by the eyes (Beatty et al., 2000)(Senior
et al., 2010)(Bremner, 2012). For the second case,
the pupil diameter signals can become a signal that
indicates the individual’s mental states, reflecting the
mental effort required to perform some task. In the
case of increased mental effort, the pupil reveals the
state of attention, receiving the ANS sympathetic sig-
nal for pupil dilation, acquiring a larger amount of vi-
sual information while performing the task, whereas
in the case of decreased cognitive effort, the pupil re-
veals the state of rest and energy saving, receiving a
parasympathetic signal for pupil constriction, acquir-
ing a smaller amount of visual information, as shown
in Figure 1 (Orsi et al., 2019).
Figure 1: Iris muscles. Adapted from (Marieb and Keller,
2011).
3 METHODS
This section describes the set of materials and meth-
ods used to perform the experiment and to process the
pupil signal, divided into 7 subsections: visual stim-
ulus; reading experiment; signal acquisition, signal
preprocessing, image preprocessing, frequency do-
main analisys and pattern recognition.
3.1 Visual Stimulus
From reports of individuals with the syndrome, the
most common visual distortions during the reading of
static texts are blurry and washout (Irlen, 2005). The
initial phase of the experiment consisted in generating
videos simulating these effects, to be later presented
to volunteers on a computer screen equipped with an
eye-tracking device. To perform this step, a code was
developed in the R language, using the image process-
ing library ”Magick” that generated a dynamic anima-
tion in .avi format that distorted cyclically from the
insertion of a text as input, intensifying and smooth-
ing within periods of time. The texts selected were
children’s stories with a low level of complexity so
that there would be no bias in less skilled individuals
or those with little grammatical knowledge.
The results were two videos, each simulating a
different visual distortion (blurry and washout) that
a voluntary person can perceive when reading a static
text, as presented in Figures 2 and 3.
The intensity of the effects was defined with a dis-
tortion parameter, ranging from 0 to 1, where 0 is the
scale with no visual distortion and 1 is the maximum
visual distortion.
3.2 Reading Experiment
The experiment included 70 participants, 30 men and
40 women, all healthy and cognitively able. In ad-
ICPRAM 2023 - 12th International Conference on Pattern Recognition Applications and Methods
510