Seeing Jesus in Toast: Neural and Behavioral Correlates of Face Pareidolia

Cognitive psychology studies the ways in which the information, obtained by human’s olfactory organs, is transmitted into brain impulses. In addition, it explores the mechanisms that are involved while the information is processed, acknowledged and stored. Moreover, this field of science strives to foresee and explain such processes as thinking, making conclusions, creating patterns, gaining experience, obtaining biases and many others, which are being generated by ongoing acquisition of information (Revlin, 2013). The similarities and patterns of individual’s perception as mental activity are studied by Gestalt psychology, being based on logically developed characteristics such as vicinity, similarity, closure, common fate symmetry, and proper continuation. Nevertheless, sometimes human brain perceives the deceptive and illusory information as genuine. Without doubt, this peculiarity of cognition evokes significant interest of scientists. In particular, in 2011 a group of researchers conducted “three behavioral experiments on the spatial characteristics evoking illusory face and letter deception” (Reith, Lee, Lui, Tian, & Huber, 2011). The results of the studies reveal that there is a relatively small number of distinguishing features of different faces. Therefore, in order to detect a human face, minimum patterns are required. In contrast to face images, one needs more related characteristics to detect letters. Furthermore, there is a decisive difference in location of the illusory matters: illusory letters must appear in the center so that one can perceive them while the faces can be detected with the offset to the left. Consequently, people may perceive illusory faces by their peripheral vision as well.

The article “Seeing Jesus in Toast: Neural and Behavioral Correlates of Face Pareidolia”discusses the continuity and progress in studying the mechanisms and peculiarities of the detection of deceptive images and letters. It reveals the experiment, conducted by Lui, Li, Feng, Li, Tian and Lee in the same field of study with the purpose to learn the peculiarities of behavioral and neural processes of face pareidolia. This term is used to name the phenomenon of delusive perception of illusory faces and/or letters.

The variables that are being examined are the process of face pareidolia; the work of the right fusiform face area (rFFA), which is responsible for the ability to recognize faces either illusory or real; the particularities of classification images and their generation. Besides, the so-called event-related potential responses (ERP) and face-specific activations belong to variables as well. The parts of human brain and those steady processes they evoke can be called predictor variables (the variables that are known and unchangeable). The process of detection of the delusive images embodies criterion variables (those that can be changed).

At the beginning of the study, the researchers make the following premises. Hypothesis 1: the perception of the illusory images (faces, letters) is reinforced by the right fusiform face area that is also responsible for recognition of real faces. Hypothesis 2: the creation of classification images is directly linked to face-specific activations. Hypothesis 3: face-specific activations function to facilitate the perception of indistinct images serving as a bridge between primary visual and prefrontal cortexes. Hypothesis 4: if the slightest resemblance to the human face is detected, this image will be advanced to the level necessary for its interpretation as a true human face. Hypothesis 5: the detection of delusive letters happens in a similar way as the perception of real ones by means of activating cortex letter-preferential regions.

The given experiment is conducted in China. The sample of the study consists of twenty Chinese, who are right-handed, healthy, young and with normal vision. The participants are native Chinese and have been living in this country their whole life. Simultaneously, for the last 10 years they have been utilizing Roman letters on a regular basis. The sample is proposed to detect the images of faces and letters in the pictures, which are divided on the basis of complexity into five levels: easy-to-detect-faces/ easy-to-detect-letters; hard-to-detect-faces/hard-to-detect-letters; and pure-noise images (Liu, Li, Feng, Li, Tian, & Lee, 2014). The experiment is divided into two parts: the aim of the first part is to train the sample to detect images. This stage is critically important because it helps to put all participants in similar external and internal conditions, thus lessening the level of stress and disorientation, or lack of understanding what exactly they are supposed to do. Including the training stage into this experiment ensures that participants’ potential to detect illusive images and their attention are supposed to be utilized to the maximum. Besides, it helps to identify and anticipate the flaws and/or difficulties of the major experiment. The main part consists of the task where the participants are asked to press the button if they are able to detect the letters or faces in the proposed pictures. It should be noted that they are warned about the difficulty level of every next set of pictures.

The results of this study reveal the following findings. As it is expected, the process of identification of the delusory faces/letters is maintained by the right fusiform face area. Besides, generation of classification images is related to face-specific activations in a way where these mechanisms of activation trigger the production of classification images. Moreover, as the researchers presume, face-specific activations establish the connection between primary visual cortex and prefrontal one. In particular, it facilitates the detection of fuzzy features. As it is expected, even if there is a minimal similarity to a human face, the mechanism of face-specific activation starts to improve the quality of this visual image. In behavioral terms, if a person is told that certain image contains the face and/or letter, this information triggers face-specific activations even if the resemblance was not detected. In other words, when an individual is encouraged to detect deceptive image, it is more probable that such will be found. Finally, the detection of illusory letters is similar to the way in which the real ones are perceived. Nevertheless, the identification of the letters that contain only vertical lines (such as i, l, x) is considerably harder comparing to those that include the curves.