minor

Author: zen-juen

paper/paper_revised.Rmd

@@ -35,7 +35,7 @@ authornote: |
 abstract: |
   Visual illusions are fascinating phenomena that have been used and studied by artists and scientists for centuries, leading to important discoveries about the neurocognitive underpinnings of perception, consciousness, and neuropsychiatric disorders such as schizophrenia or autism. Surprisingly, despite their historical and theoretical importance as psychological stimuli, there is no dedicated software, nor consistent approach, to generate illusions in a systemic fashion. Instead, scientists have to craft them by hand in an idiosyncratic fashion, or use pre-made images not tailored for the specific needs of their studies. This, in turn, hinders the reproducibility of illusion-based research, narrowing possibilities for scientific breakthroughs and their applications. With the aim of addressing this gap, *Pyllusion* is a Python-based open-source software (freely available at https://github.com/RealityBending/Pyllusion), that offers a framework to manipulate and generate illusions in a systematic way, compatible with different output formats such as image files (.png, .jpg, .tiff, etc.) or experimental software (such as *PsychoPy*).
 keywords          : "Pyllusion, Visual Illusions, Python, Software, Open-source, Perception"
-wordcount         : "5348"
+wordcount         : "5375"
 bibliography      : ["paper.bib"]
 floatsintext      : yes
 figsintext        : yes
@@ -70,12 +70,12 @@ library("papaja")
 
 <!-- Importance and use of visual illusions-->
 
-Visual illusions have been observed for hundreds of years [@LuckieshVisualIllusions1965], many of which were described in print [@helmholtz1856handbuch]. In general terms, a visual illusion can be thought of as the inaccurate perception of a visual stimulus or a given attribute, be it geometrical (size, shape, or angle), or another property such as colour [@muller1896lehre; @howe2005muller; @delboeuf1893nouvelle; @ebbinghaus1902grundzuge; @roberts2005roles; @adelson200024]. Often, an illusory perception resists 'correction' in perception even after an observer has been made aware of the misperception. Novel examples of illusions are still observed and have even cropped up on social media platforms, a famous example being 'The Dress Illusion' [as discussed by @schlaffke2015brain], which some people perceive as white and yellow, whereas others as black and blue - this is thought to illustrate how perceptual priors (i.e., expectations regarding lighting conditions) can bias our conscious representation of an object. See @ninio2014geometrical, @LuckieshVisualIllusions1965, and @robinson1972psychology for extensive collections of visual illusions. Overall, these illusions show how our phenomenological experience is critically shaped by contextual information and prior expectations.
+Visual illusions have been observed for hundreds of years [@LuckieshVisualIllusions1965], many of which were described in print [@helmholtz1856handbuch]. In general terms, a visual illusion can be thought of as the inaccurate perception of a visual stimulus or a given attribute, be it geometrical (size, shape, or angle), or another property such as colour [@muller1896lehre; @howe2005muller; @delboeuf1893nouvelle; @ebbinghaus1902grundzuge; @roberts2005roles; @adelson200024]. Often, an illusory perception resists 'correction' in perception even after an observer has been made aware of the misperception. Novel examples of illusions are still observed and have even cropped up on social media platforms, a famous example being 'The Dress Illusion' [as discussed by @schlaffke2015brain], which some people perceive as white and yellow, whereas others as black and blue - this is thought to illustrate how perceptual priors (i.e., expectations regarding lighting conditions) can bias our conscious representation of an object. See @ninio2014geometrical, @LuckieshVisualIllusions1965, and @robinson1972psychology for extensive collections of visual illusions. Overall, these illusions show how our phenomenological experience is critically shaped by contextual information and prior expectations. In the next few sections, we discuss some important conceptual and methodological issues in illusion science and how our software, *Pyllusion*, facilitates systematic investigations of these matters.
 
 Entertainment value aside, illusions have had considerable importance in the history of psychological science. Visual illusions have helped scientists understand the architecture of the eye and its relationship with processes and structures involved further up stream in the brain, the dynamic interaction of these processes, and visual coding in the brain in general [@carbon2014understanding; @forte2005inter; @clifford2002perceptual]. Illusions such as those associated with colour perception, orientation perception, and motion perception, have all been informative of neuronal activity/processes both at the level of the eye and the brain via their measurement [@webster1996human; @witkin1948studies; @mackay1957moving; @Holland1965; @curran2009hierarchy]. Visual illusions, and perceptual illusions more generally, are a powerful tool in human perception and brain research, which in turn can inform artificial cognitive systems design considerations [@carbon2014understanding; @boyce2020optimality]. Beyond low-level perceptual mechanisms, illusions can also be powerful tools to understand higher-order processes related to phenomenal consciousness [@mahon2018role], as well as neurocognitive disturbances.
 
 <!-- Psychopathology-->
-Illusory paradigms have been widely used to investigate neurocognitive deficits as visual illusions highlight the influence of context on visual perception [@corbett2006observer; @chen2015contextual; @roberts2005roles]. Visual illusions are thus valuable tools for investigating core features of pathological conditions, such as atypical integration processes in schizophrenia [@clifford2014tilt; @thakkar2021; @palmer2018perceptual; @notredame2014visual; @king2017review; @liddle1987schizophrenic; @dakin2005weak; @tibber2013visual] and in autistic spectrum disorder (ASD) [@mitchell2010susceptibility; @walter2009specific; @gori2016visual]. Evidence from visual illusions research has garnered substantial support for an account - the Predictive Coding framework [@friston2009predictive] - which posits that illusory perception typically arises because of a strong systematic bias for prior beliefs (top-down influence) that are mismatched with actual sensory evidence, causing the generation of an objectively wrong but more plausible percept [e.g., two objectively equivalent-sized circles being interpreted as different sizes because of their surrounding context, @notredame2014visual]. A greater resistance to visual illusions (such as that observed in some pathological conditions) is then interpreted as a product of reduced adaptive top-down influence [@schneider2002reduced; @koethe2009binocular] and an over-reliance on sensory evidence [i.e., bottom-up processes, @dima2010impaired]. This all helps to underscore the neurocomputational mechanisms that are fundamental to psychiatric and psychological disorders [@sterzer2018predictive].
+Illusory paradigms have been widely used to investigate neurocognitive deficits as visual illusions highlight the influence of context on visual perception [@corbett2006observer; @chen2015contextual; @roberts2005roles]. Visual illusions are thus valuable tools for investigating core features of pathological conditions, such as atypical integration processes in schizophrenia [@clifford2014tilt; @thakkar2021; @palmer2018perceptual; @notredame2014visual; @king2017review; @liddle1987schizophrenic; @dakin2005weak; @tibber2013visual] and in autistic spectrum disorder (ASD) [@mitchell2010susceptibility; @walter2009specific; @gori2016visual]. Evidence from visual illusions research has garnered substantial support for an account - the Predictive Coding framework [@friston2009predictive] - which posits that illusory perception typically arises because of a strong systematic bias for prior beliefs (top-down influence) that are mismatched with actual sensory evidence. This in turn causes the generation of an objectively wrong but more plausible percept [e.g., two objectively equivalent-sized circles being interpreted as different sizes because of their surrounding context, @notredame2014visual]. A greater resistance to visual illusions (such as that observed in some pathological conditions) is then interpreted as a product of reduced adaptive top-down influence [@schneider2002reduced; @koethe2009binocular] and an over-reliance on sensory evidence [i.e., bottom-up processes, @dima2010impaired]. This all helps to underscore the neurocomputational mechanisms that are fundamental to psychiatric and psychological disorders [@sterzer2018predictive].
 
 <!-- Replicability issues on parameter manipulation -->
 The susceptibility to illusions can be influenced by manipulating specific stimulus features. For instance, inverting the orientation of the illusion (by 180 degrees) can influence perceptions of the Contrast and Ponzo illusions where in both cases, illusion magnitude is greater when the illusion presentation is in its 'upright' position [@poom2020influences]. Other feature manipulations, such as those related to the lengths and lightness contrast of distractor lines, have been demonstrated to modulate illusion magnitude for both the Ponzo illusion [@jaeger1980effect] and the Müller-Lyer illusion [@jaeger1980effect; @wickelgren1965brightness; @jaeger1975effect; @restle1977size] in similar ways. However, varying stimulus features does not always produce consistent results in terms of the perceived illusion. For instance, in the Ebbinghaus illusion, it is unclear whether increasing the number of small surrounding context circles increases or decreases the perceived size of the target circle [@girgus1972interrelationship; @massaro1971judgmental; @jaeger1978ebbinghaus]. This has led to competing theories and different predictions [@woloszyn2010contrasting]. In this regard, the ability to experimentally manipulate various parameters will be crucial for testing these theories, in turn deepening our understanding of the distinct neurocognitive mechanisms that can be inferred from the systematic effects these parameters have on illusion perception [e.g., the role of expectations and familiarity, @poom2020influences], as well as the neural basis of psychopathology [@tibber2013visual; @parnas2001visual; @yang2013visual; @spencer2014oscillatory].