Acquired Prosopagnosia: Seeing Words Without Faces
Wildman
Susilo, T., Wright, V., Tree, J. J., & Duchaine, B. (2015). Acquired prosopagnosia without word recognition deficits. Cognitive Neuropsychology, 32, 321-339. doi:10.1080/02643294.2015
.1081882
Acquired Prosopagnosia: Seeing Words Without Faces
Part 1: Acquired Prosopagnosia: An Overview of Face Blindness as a Discrete Syndrome
Prosopagnosia is a cognitive disorder, impacting the ability to recognize familiar faces. Unlike reading, facial recognition is not taught, and unlike memory, it is not typically tested. Rather, facial recognition is important, yet relatively unnoticed skill unless there is an occurrence of prosopagnosia, or face blindness. Prosopagnosia may manifest on a continuum, with the sufferer unable to identify his or her face, or with limited recognition of certain people or expressions. In addition, Prosopagnosia may occur as one symptom of other neurological issues, such as Alzheimer’s disease (Barton, 2008). Prosopagnosia can be developmental, from birth, or can be acquired from a brain injury. In the latter, lesions are often found in the fusiform gyri and the lingual of the medial occipitotemporal cortex, and at times in the anterior temporal lobe (Barton, 2003; Deen, 2010). Lesions may be bilateral or unilaterally on the right side (Barton, 2003). Whereas prosopagnosia may occur with other visual agnosias, such as object identification agnosia, or word agnosia (Ogden, 2005), historically, prosopagnosia was deemed a syndrome that may be discrete (Duchaine & Nakayama, 2005).
However, recent theories have linked prosopagnosia to word-recognition deficits, challenging the widely held belief that prosopagnosia may occur independently of other agnosias (Behrmann & Plaut, 2013). This leads to the major issue addressed in the present paper: Can acquired prosopagnosia occur as a discrete syndrome, or is prosopagnosia linked to word-recognition deficits? This paper is based upon an analysis of Susilo, Wright, Tree, and Duchaine’s (2015) article, “Acquired Prosopagnosia Without Word Recognition Deficits.” Susilo et al.’s study found that acquired prosopagnosia may occur without word deficit, thus challenging a theoretical paper by Behrmann and Plaut (2013), which linked face and word recognition through a common circuit. The many-to-many theory (MTMT) asserts that visual recognition of faces and words do not occur through independent mechanisms, rather, prosopagnosia is linked to word recognition deficits through overlapping, nondiscrete neural circuits (Behrmann & Plaut, 2013).
To understand Susilo et al.’s (2015) hypothesis that face blindness may occur without written word agnosia, prosopagnosia and related agnosia literature have been reviewed, including (a) Behrmann and Plaut’s (2013) theory of connected pathways, (b) a study that assessed word-recognition deficits in a developmental prosopagnosia subject (Rubino, Corrow, Corrow, Duchaine, & Barton, 2016), and (c) two studies that analyzed prosopagnosia and the relationship to object agnosia (Duchaine & Nakayama, 2005; Farah, Levinson, & Klein, 1995).
Next, a critical analysis of Susilo et al.’s article evaluates their experimental question, to determine if prosopagnosia occurs with or without word-recognition deficits. This analysis details the five prosopagnosia patients in Susilo et al.’s study, the test methods, and results. Susilo et al.’s finding that prosopagnosia may be a discrete agnosia, challenging Behrmann and Plaut’s theory that word and facial recognition occur through common mechanisms, was analyzed. Finally, I propose an experiment, which builds on the findings of Susilo et al. (2015), as well as Behrmann and Plaut (2013), testing subjects with word-recognition deficits for prosopagnosia.
Literature review
To further assess the theory that prosopagnosia is a discrete or nondiscrete condition (Behrmann & Plaut, 2013), a word-recognition deficit study (Rubino et al., 2016) and object recognition agnosia studies (Duchaine & Nakayama, 2005; Farah et al., 1995), were reviewed. This paper will address the debate as to whether prosopagnosia is connected or dissociable from a word-recognition deficit through a distributed but integrated system (Behrmann & Plaut, 2013).
Seeing Faces and Words: Connected Pathways?
The integrated MTMT postulates that visual recognition of faces and words occurs within common mechanisms in the brain (Behrmann & Plaut, 2013), linking prosopagnosia to word-recognition deficits. In a theoretical paper based upon imaging, prior behavioral research, and computational modeling, Behrmann and Plaut (2013) asserted that the brain is organized not through circumscribed centers, but through distributed circuits. The theory suggests visual ability is through an integrated network within a cortical region, not a discrete area of the brain. This challenges beliefs that facial recognition is an independent function, located in the opposite hemisphere of the brain from the word-processing mechanism, which also functions independently. The authors proposed the brain has overlapping neural pathways; where the right side of the brain plays a dominant role in face recognition, and the left side of the brain plays a dominant role in word recognition, and the sides are not mutually exclusive.
Prior research evaluated brain hemisphere dominance in adults, adolescents, and children (Dundas, Plaut, & Behrmann, as cited in Behrmann & Plaut, 2013). Adolescents presented no field of dominance for faces and only right-side dominance for words. Children’s results were consistent with adolescents, albeit with lower accuracy. Adults showed the right and left field dominance for word and face discrimination, respectively. The authors suggested face recognition developed with reading ability, supporting the theory that visual and reading acuity are not seperate. Yet, Susilo et al. (2015) did not test subjects with alexia for prosopagnosia or test prosopagnosia with visual agnosias.
Word Recognition in Developmental Prosopagnosia
In Rubino et al.’s (2016) study, they sought to understand the impact on word recognition with developmental prosopagnosia patients. Ten adults with developmental prosopagnosia and a control group performed two tests: (a) a word-length effect test and (b) a word-versus-text-style test. In the word-length test, subjects sorted words from 3 to 9 letters on a screen and were asked to read the word aloud as they were recorded. In the sorting by word versus text style test, 10 words, handwritten by 10 different subjects, with word length from 11 to 2 letters, were sorted (Rubino et al., 2016). The same words were created using various computer fonts. Subjects were asked to sort the words into discrete piles, to test the impact of handwriting cognition. The developmental prosopagnosia group performed within norms for word-length effect. There was no difference between the developmental prosopagnosia group on average compared to the control group sorting by word style, in completion times or accuracy. The results did not support the MTMT of common pathways, impacting both face and word recognition.
Face Blindness and Object Impairment
Previous research by Farah et al. (1995) investigated whether people with acquired prosopagnosia might have object impairment. Their study tested a subject with prosopagnosia in comparison to 10 control subjects without prosopagnosia using behavioral tests. L.H., a 37-year-old man, suffered severe prosopagnosia, unable to recognize even his own face, due to head injuries from an automobile accident and related surgery. This impacted temporal-occipital regions on both sides, the anterior temporal lobe and right interior frontal lobe, as demonstrated in a neurological exam, and a computed tomography (CT) scan. L.H. had returned to college; and could read (Farah et al., 1995). The first test was to determine memory of an assortment of ordinary objects (e.g., a chair) and photographs of faces (Farah et al., 1995). The second experiment challenged the notion that facial recognition is more difficult due to the similarities, whereby subjects evaluated black-and-white photographs of clean-shaven men without glasses and photographs of eyeglasses. In both studies, the control group scored significantly better than L.H. in facial recognition. In the first test, L.H. performed significantly better at object recognition, whereas the control group scored the same on face and object testing. In the second test, the control group performed significantly worse in eyeglass than facial recognition. L.H. scored significantly lower on faces relative to eyeglasses when compared to the control group, further confirming his face recognition was more impaired than the control group’s (Farah et al., 1995).
Duchaine and Nakayama (2005) evaluated seven developmental prosopagnosia adults compared to a normal control group. Their objective was to determine if prosopagnosia is discrete from object agnosia. The authors administered old–new recognition memory tests, using stimuli such as faces, horses, houses, tools, cars, guns, and natural scenes as objects. Participants completed the recognition memory tests with varying combinations of stimuli type. Of the seven subjects, four had only prosopagnosia deficit, with significantly better scores for object compared to facial recognition. The other three subjects appeared to have deficits in prosopagnosia and object agnosia. Reaction times were also tested with the theory that prosopagnosia patents may spend more time on other object viewing to compensate for face blindness, contributing to better results (Gauthier, Behrmann, and Tarr as cited in Duchaine & Nakayama, 2005). Prosopagnosia subjects spent more time on object-recognition tasks, but also spent more time on facial-recognition tasks, with significantly poorer results in face recognition versus the control group and compared to their own object recognition (Duchaine & Nakayama 2005). Both studies (Duchaine & Nakayama 2005; Farah et al., 1995) indicated prosopagnosia without object agnosia. The authors of each study surmised prosopagnosia may be distinct from object agnosia. These findings further support that acquired prosopagnosia may be a discrete syndrome.
In summary, Behrmann and Plaut’s (2013) theory was an attempt to link prosopagnosia to word-recognition deficits, based upon imaging and past behavioral research, but lacked studies with prosopagnosia patients without word-recognition deficit or object agnosia. Behrmann and Plaut also did not review developmental prosopagnosia subjects. Rubino et al. (2016) further supported the theory that prosopagnosia is discrete from word-recognition deficits, as developmental prosopagnosia subjects did not demonstrate word-recognition deficits when tested. Studies by researchers such as Farah et al. (1995) and Duchaine and Nakayama (2005) supported the theory that prosopagnosia may occur without object impairment, providing an additional premise to support the hypothesis that prosopagnosia may occur without word-recognition deficits.
Part 2: Prosopagnosia Without Word-Recognition Deficits
The goal of Susilo et al.’s (2015) study, “Acquired Prosopagnosia Without Word Recognition Deficits” was to test the validity of the MTMT by evaluating whether patients with prosopagnosia also demonstrated word-recognition deficits. This theory acknowledged that word-and-face recognition function is lateralized, but not fully discrete (Behrmann & Plaut, 2013). The MTMT suggests that distributed circuits control visual recognition; rather than circumscribed centers. The MTMT posits that prosopagnosia patients also have word-recognition deficits (Behrmann & Plaut, 2013). To test this theory, Susilo et al. tested five patients with acquired prosopagnosia to determine if they also demonstrated word-recognition agnosia.
Study Participants
Five patients with varying degrees of acquired prosopagnosia with no other cognitive abnormalities were identified both symptomatically through three tests, and with magnetic resonance imaging (MRI) to confirm brain lesions (Susilo et al., 2015). The patients were identified through registration on the website, faceblind.org. The three tests used to verify prosopagnosia were (a) Old–New Face Recognition Test (Duchaine, Yovel, Butterworth, & Nakayama, 2006), (b) the Famous Face Test (Duchaine & Nakayama, 2005), and (c) the Cambridge Face Memory Test (Duchaine & Nakayama, 2006). All five patients demonstrated prosopagnosia on the three tests compared to control norms. The MRI structural scans of the five patients were used to verify brain lesions. The five acquired prosopagnosia patients included one younger subject, a 31-year-old man, and five older subjects, three women aged 50, 52, and 64 and one 57-year-old man. Although study participants all tested positive for prosopagnosia, the subjects varied in when and how the brain injury that caused the face blindness occurred. None of the subjects had experienced prosopagnosia prior to brain injury. Specific descriptions of each participant’s condition, are below.
Galen. A 31-year-old, right-handed male, Galen was working as a physician in a hospital. A craniotomy for an arteriovenous malformation in his right temporal lobe rendered Galen with prosopagnosia. A temporary left-superior quadrantanopia appeared to have resolved, confirmed via testing with other visual cognition testing as normal (Susilo et al., 2015).
Faith. A 50-year-old, right-handed woman, working as a teacher, Faith had a right occipitotemporal resection for epilepsy in 2012, resulting in prosopagnosia. Faith has difficulty even recognizing immediate family and cannot identify gaze discrimination or facial expressions. Faith’s general visual abilities are normal. (Susilo et al., 2015).
Kili. A 52-year-old, right-handed woman, Kili was working as a freelance writer with a right occipital lobe infarction following a stroke causing prosopagnosia. Following the stroke, identifying family became difficult. Using the Humprey visual perimetry, Kili was diagnosed with a left hemianopia (Susilo et al., 2015).
Lily. A 64-year-old, right-handed woman, Lilly was working in health care. Surgery for arteriovenous fistula resulted in prosopagnosia. MRI testing indicated a stroke postoperatively, with material used to repair the fistula affixed to the adjacent artery, creating a lesion on the right ventral visual pathway, impeding the right fusiform gyrus. (Susilo et al., 2015).
Herschel. A 57-year-old, right-handed man, Herschel was working in technology, with a degree in astronomy. His acquired prosopagnosia was caused by a stroke in 2008, with additional visual issues of navigation and an upper left quandrantanopia. Herschel suffered a second stroke 4 months later, which created an upper right quandrantanopia. A MRI confirmed bilateral occipitotemporal lesions, with right hemisphere increased dominance (Susilo et al., 2015).
Control Groups
Susilo et al. (2015) included two control groups, one for the younger participant and one for the older participant group. The younger acquired prosopagnosia subject, the 31-year-old male, was tested in comparison to a control group of 14 college students, five males and nine females, eight located in the United States at Dartmouth College and six located in the United Kingdom, attending University of Swansea and University of Aberystwyth. They ranged in age from 20–33 (M=23.2, SD=4.1 years). The four older acquired prosopagnosia participants (three female, age 50, 52, and 64, and one male, age 57) were compared to a control group of 12 similar-age adults, four males and eight females, with ages ranging from 56–66 (M=62.3, SD=3.1 years). The participants in the older control group participated in a multiple-choice, 20-question vocabulary test to verify a vocabulary similar to the test group, due to deviation in vocabulary for older adults. The older control group’s mean accuracy was similar to the older prosopagnosia group, verifying a normal-range and consistent working vocabulary in test compared to control groups.
Methods and Experiments
Various behavioral experiments totaled 1,200 trials, with seven word-recognition tasks, including five reading tests and two lexical decision tests, analyzed used Crawford’s t test (Susilo et al., 2015). Participant tests were conducted in a random order. Variables included word length, the age when a word was learned (the age of acquisition), and words more and less frequently used. Specific experiment descriptions and details, follow.
Lexical decision task: Frequency x Age of Acquisition. This test measured recognition of early and late acquired words, and high and low-frequency words, by asking subjects to identify words in comparison to nonwords. Each subject viewed 40 words acquired early and 40 words acquired late (Bristol Norms, Stadthagen-Gonzalez, & Davis, as cited in Susilo et al., 2015), and 40 high- and 40 low-frequency words (selected from CELEX database; Bayern, Piepenbrock, & Van Rijn, as cited in Susilo et al., 2015). Words and nonwords were paired through factors including orthographic neighbors, string length, and frequency of biagram. The test was administered on a screen that started with a 200 ms view of a fixation cross. The stimuli tested were then placed at the screen’s target location. The font used for testing was Arial, 24-point, lower case, black type on a white screen. Subjects were given six-word and six-nonword practice trials. In the test, subjects were asked to determine with correctness and speed, whether the “word” presented was a word or a nonword by pressing a specific key on the keyboard. Once a response was entered, a lit asterisk appeared for 500 ms. At that point, the target cross appeared again for 2000 ms as a signal for the next event to start.
Lexical decision task: Length. This test, again, measured word recognition through sorting of words compared to nonwords. This time, the words were divided into word lengths of 7, 5, or 3 letters. This test compared 120 words and nonwords, with 40 of each length. Once again, ARC Nonword Database created the nonwords. The protocol for testing was the same as the previous test, with participants asked to identify the words and the nonwords through pushing one of two buttons on the keyboard (Susilo et al., 2015).
Reading aloud task: Frequency x Age of Acquisition. This test measured participants’ ability to read aloud low- and high-frequency words as well as early and late acquired words. Font was consistent with the prior test. Six practice trials were given. The target cross was used in the center of the screen, for 2000 ms, after which the word tested appeared. The subjects then read each word, measured for correctness and speed. A new word would appear on the screen after the subject verbally responded. After a response, an asterisk appeared for 500-ms, then the target cross for an additional 2000 ms prior to the next word appearing. Responses were checked for accuracy through use of a digital voice recorder and verified (Susilo et al., 2015).
Reading aloud task: Length. This test measured the ability to correctly and quickly read aloud words with different lengths. Words of 7, 5, and 3 letters were divided into three groups of 40 words, with 120 total words. Protocol was consistent with the prior test (Susilo et al., 2015).
Reading aloud task: Confusability. This task included three parts; average, summed and N. confusability testing. First, the average confusability consisted of reading aloud 120 words. Words were organized by average letter confusability, N, and frequency. The protocol was consistent with the prior test (Susilo et al., 2015). The summed confusability test measured reading aloud 120 words based on summed letter confusability (Susilo et al., 2015). Finally, the N confusability test measured reading aloud 200 4-letter words, including 50 words that were low confusability/low N, 50 words that were low confusability/high N, 50 words that were high confusability high N, and 50 words that were high confusability/low N. Protocol was consistent with the prior test (Susilo et al., 2015).
Results
Four of five of the prosopagnosia patients consistently scored within norms in comparison to the control group for the Susilo et al. (2015) study. This suggested no word-recognition agnosia for the majority of the prosopagnosia subjects, with Galen, Faith, Herschel, and Lily demonstrating normal results for word recognition. Kili demonstrated word-recognition deficits, with below normal range for speed on almost all tests. Kili also demonstrated more erroneous answers compared to controls for low-frequency words. Galen scored normal in all tasks. Faith, Herschel and Lily each had a score out of norm in a limited number of comparisons, albeit within predicted norms. As Susilo et al. suggested, the average projected to be abnormal in a study of 56 statistical tasks would be 2.8, at an alpha of .05. Kili was out of the norm, with 22 tasks demonstrating results poorer than the controls. On word-length effect, out of four different tests of slopes, Kili, Lily, and Faith showed one abnormal slope each. Although their ranges from 16 to -38ms/letter were statistically abnormal compared to the slope, the scores were within the -6 to 32 ms/letter range, similar to readers without word-recognition deficits (Barton, Hanif, Eklinder, Bjornstorm, & Hills, as cited in Susilo et al., 2015). Specific results by task follow.
Lexical decision task: Frequency x Age of Acquisition. Four out of five acquired prosopagnosia subjects performed within norms in identifying words and nonwords. Kili was the exception, with a higher error rate for low-frequency words (Susilo et al., 2015).
Lexical decision task: Length. Four out of five subjects identified words and nonwords within norms, with Kili as the exception, significantly slower for 7- and-5-letter words and with more errors for 7-letter words than controls. The word-length effect, analyzing by-letter reading for pure alexia, was also measured through a regression of response time versus the number of letters. All subjects other than Faith tested normal compared to control. (Susilo et al., 2015).
Reading aloud task: Frequency x Age of Acquisition. Results were mixed. Two subjects, Galen and Lily, performed within norms. Herschel and Faith demonstrated a higher error rate than the control group with low-frequency/early-acquisition words. Kili demonstrated slower reading for low-frequency/early-acquisition words and had higher error rates for all but high-frequency/late acquisition words than the control groups. (Susilo et al., 2015).
Reading aloud task: Length. Four out of five subjects performed normal for reading-aloud length. Kili tested slower than norms for most of this test with a higher error rate than the control group for 5-letter words. In the word-length testing effect, three of five subjects tested normal, with Kili and Lily demonstrating a slower slope. (Susilo et al., 2015).
Reading aloud task: Confusability. All subjects tested in the normal slopes for average confusablity. Four out of five subjects tested normal reading aloud in relation to average confusability. For summed letter confusability, all subjects performed in the normal slopes. Four out of five subjects demonstrated normal reading aloud in relation to summed confusability. For N confusability, four out of five subjects performed normally. Kili performed slower for most tasks than the control group, with low-N/low-confusability words (Susilo et al., 2015).
Conclusion. Outcomes from Susilo et al.’s (2015) study indicated prosopagnosia can be a discrete syndrome from word-recognition deficit, with no association between facial and word recognition, inconsistent with Behrmann and Plaut’s (2013) MTMT. Although Kili demonstrated word-recognition deficits, Susilo et al. hypothesized this might be due to the presence of the hemianopia, which has associations with alexia (Barton et al., as cited in Susilo et al., 2015). Whereas Kili’s results are important, they are the outlier, as the four other subjects with prosopagnosia demonstrated normal word recognition on a comprehensive battery of tests.
This finding is also consistent with the literature review, as the developmental prosopagnosia and word-recognition deficits study by Rubino et al. (2016) showed no relationship between the two agnosia, indicating prosopagnosia is dissociable from word-recognition deficits. Studies regarding prosopagnosia and the link to object disorder also indicated prosopagnosia as a discrete condition (Duchaine & Nakayama, 2005; Farah et al., 1995), further supporting findings from Susilo et al. (2015). Based upon study findings, there is no association between facial and word recognition, thereby failing to support predictions in Behrmann and Plaut’s (2013) MTMT.
Critical Commentary
Strengths of Susilo et al.’s (2015) study included a robust methodology, in excess of 1200 trials of word recognition testing among acquired prosopagnosia subjects. This included seven word recognition tests. Additionally, the authors elected to increase the probability of finding subtle deficits by not correcting for multiple comparisons, thereby increasing the ability to detect some form of word-recognition deficits. The study indicated prosopagnosia without word deficit, despite this heightened and robust test methodology, with four out of five patients testing positive for acquired prosopagnosia, yet negative for word-recognition deficits.
However, the authors tested a limited subject population in both the test and control groups. The test group was limited in size, age, and type of agnosia. Only five subjects were tested, two men and three women, and only one subject, Galen, age 31, was in the younger group. All other subjects were 50 years of age or older (Susilo et al. 2015). Prosopagnosia patients were limited to acquired prosopagnosia patients; none had developmental prosopagnosia. All subjects worked in a professional field and all in the control group either attended or worked at the university, thereby educational levels might not represent the general population. In addition, no patients with alexia were included (Susilo et al. 2015), the other hypothesis from the Behrmann and Plaut (2013) study.
Part 3: Prosopagnosia Further Testing
Research Objectives, Question, and Hypotheses
The primary objective of the proposed research is to determine whether word and face recognition share a common pathway, whereby patients with word-recognition deficits also demonstrate prosopagnosia. Susilo et al.’s (2015) study focused exclusively on whether patients with acquired prosopagnosia also demonstrated word-recognition deficits. However, their study did not include research with alexia patients for prosopagnosia. The question this proposed research will attempt to answer is the following: Do patients with word-recognition deficits also demonstrate prosopagnosia? Therefore, the study will test Behrmann and Plaut’s (2013) MTMT from the inverse approach of Susilo et al.’s study. Patients will be selected with alexia, and tested on their ability to identify faces.
The hypotheses are as follows:
1) Patients with alexia will not consistently demonstrate prosopagnosia, consistent with Susilo et al.’s (2015) study findings that posit prosopagnosia and alexia lack overlapping brain mechanisms.
2) Alternatively, patients with alexia will consistently demonstrate prosopagnosia, consistent with Behrman and Plaut’s (2013) theory of overlapping circuits in the brain impacting the cognition of both facial and word recognition.
Background and Rationale
Currently the scientific community debates the MTMT for prosopagnosia and the relationship of prosopagnosia with word-recognition deficits. Susilo et al. (2015) tested prosopagnosia patients but did not test alexia subjects. Behrmann and Plaut (2013) cited research that indicated subjects lacking formal education exhibited increased facial cognition in the left hemisphere (Dehaene, S., et al. 2010, as cited in Behrmann & Plaut 2013). This supports the theory that word and face recognition cognition are not fully discrete (Behrmann & Plaut, 2013). Therefore, this proposed study of alexia patients and the link to prosopagnosia is important, to further support or challenge the MTMT.
Method
To achieve my objectives, I will conduct both within-subject and between-subject tests. The within-subject experiment will test a group of alexia-screened subjects and will compare within-subject group facial-recognition abilities. The between-group testing will test the group of alexia patients in comparison to a control group without alexia, to determine word-recognition and facial-recognition deficits. Testing will be conducted among both alexia and control groups to determine (a) word-recognition deficits between alexia-screened subjects and the non-alexia controls, (b) face recognition tested within alexia patients, and (c) and face-recognition experiments between alexia patients compared to the control group.
Instruments. The test methods will include the three word-recognition behavioral tests and the seven word-recognition tests used by Susilo et al. (2015), through the inverse lens, testing alexia patients for prosopagnosia. The same test methods those researchers used to test prosopagnosia patients for both prosopagnosia and for word-recognition deficits will be used for alexia patients to test for prosopagnosia, as, the use of the test methods has been qualified. Further, use of the same test methods allows for comparability between studies. The descriptions and rationale of the facial recognition testing are listed below. The seven word-recognition tests were already explained in a previous section of this paper, as this battery was described in the Susilo et al. (2015) Therefore, details of the word-recogntion tests will not be repeated.
Prosopagnosia tests. Consistent with Susilo et al.’s (2015) study, three tests will be used for determination of prosopagnosia: (a) Old–New Face Recognition Test (Duchaine, et al., 2006; Farah et al. 1995), (b) the Famous Face Test (Duchaine & Nakayama, 2005), and (c) the Cambridge Face Memory Test (Duchaine & Nakayama, 2006).
Old-New test. The old-new type of test has been used to verify agnosia for years. Farah et al. (1995) successfully used the test to verify prosopagnosia patients. Subjects were shown black-and-white pictures of faces and objects. They are allowed to study the items and told they will later see a similar picture that they will need to identify. The test indicated showed facial recognition was deficient to object recognition (Farah et al., 1995). Duchaine et al. (2006) further used this methodology to verify prosopagnosia in comparison to object agnosia, with success.
Famous Face Test. This test has been used for a variety of memory tests through the patient’s ability to identify famous people in photographs. Subjects view black-and-white pictures of famous people in popular culture. Duchaine and Nakayama (2005) used the Famous Face Test to validate developmental prosopagnosia.
Cambridge Face Memory Test. Duchaine and Nakayama (2006) developed this test specifically to improve the diagnosis of prosopagnosia. The test challenges facial recognition as subjects are shown six male target faces, which subjects must identify throughout the test, intermixed within three faces, at various angles and with novel images introduced. This test includes 72 images in total, with average scores of 58; with prosopagnosia subjects, average score is 37 (Duchaine and Nakayama, 2006). This test has been qualified for identification of prosopagnosia and is appropriate for this study.
Patient group. One group of 10 alexia patients will be recruited, ages 18-65, evenly divided between men and women, with a control group with similar age and gender to the alexia patients. Although different than Susilo et al, (2015), who recruited two age groups, ages 18-34 and 35-65, Susilo et al. also found no significant difference in the outcomes for the younger participant compared to the older subjects. Alexia patients will be confirmed via scans, typically via left hemisphere lesions, and through the seven-test series of word-recognition testing used by Susilo et al.
Expected Results.
For the first hypothesis, the expected results are that (a) Patients with alexia will be able to recognize faces consistently, and therefore display no symptoms of prosopagnosia, consistent with the hypothesis, and (b) the facial recognition slope will be consistent to the control group; and significantly higher than the word-recognition slope amongst alexia patients.
For the second, alternative hypothesis, the expected results are that: (a) patients with alexia will be unable to recognize faces consistently and therefore display symptoms of prosopagnosia, consistent with the alternative hypothesis, and (b) the facial-recognition slope will be significantly lower than that of the control group.
Graphs or figures showing potential results if you wish (not required) THOUGHTS?
Conclusion
Studies have postulated prosopagnosia and the link to alexia, through common brain mechanisms (Behrmann & Plaut, 2013). Susilo et al. (2015) conducted testing through the lens of the prosopagnosia patient. The study of the MTMT, that the brain dominance by hemisphere is not fully discrete, but rather shares connections, would be furthered by tests with alexia patients for prosopagnosia. Should this proposed study build upon Susilo et al.’s findings, further invalidating the MTMT, rather than continue to pursue the theory of connected pathways, new theories should be pursued. However, if alexia patients are proven to exhibit prosopagnosia, this finding may lead to further work on the MTMT. Susilo, et al. submitted that pure alexia subjects in many cases have varying degrees of some form of facial recognition deficits. Therefore, this study is important for continued testing of the MTMT.
References
Behrmann, M. M., & Plaut, D. C. (2013). Distributed circuits, not circumscribed centers, mediate visual recognition. Trends in Cognitive Sciences, 17, 210-219. doi:10.1016/j.tics.2013
.03.007
Barton, J. J. (2003). Disorders of face perception and recognition. Neurologic Clinics, 21, 521-548. doi:10.1016/S0733-8619(02)00106-8
Barton, J. J. (2008). Structure and function in acquired prosopagnosia: Lessons from a series of 10 patients with brain damage. Journal of Neuropsychology, 2(1), 197-225. doi:10
.1348/174866407X214172
Deen, B. (2010, February 25). Prosopagnosia—Whose face is it? Yale Scientific. Retrieved from http://www.yalescientific.org/2010/02/prosopagnosia-whose-face-is-it/
Duchaine, B. B., & Nakayama, K. (2005). Dissociations of face and object recognition in developmental prosopagnosia. Journal of Cognitive Neuroscience, 17, 249-261. doi:10
.1162/0898929053124857
Duchaine, B., & Nakayama, K. (2006). The Cambridge Face Memory Test: Results for neurologically intact individuals and an investigation of its validity using inverted face stimuli and prosopagnosic participants. Neuropsychologia, 44, 576-585.
Duchaine, B., Yovel, G., Butterworth, E., & Nakayama, K. (2006). Prosopagnosia as an impairment to face-specific mechanisms: Elimination of the alternative hypotheses in a developmental case. Cognitive Neuropsychology, 23, 714-747. doi:10.1080
/02643290500441296
Farah, M., Levinson, K., & Klein, K. (1995). Face perception and within-category discrimination in prosopagnosia. Neuropsychologia, 33, 669-667. doi:10.1016/0028-3932(95)00002
Ogden, J. (2005). Fractured minds: A case-study approach to clinical neuropsychology (2nd ed.). Oxford, England: Oxford University Press.
Susilo, T., Wright, V., Tree, J. J., & Duchaine, B. (2015). Acquired prosopagnosia without word recognition deficits. Cognitive Neuropsychology, 32, 321-339. doi:10.1080/02643294
.2015.1081882
Rubino, C., Corrow, S. L., Corrow, J. C., Duchaine, B., & Barton, J. S. (2016). Word and text processing in developmental prosopagnosia. Cognitive Neuropsychology, 33, 315-328. doi:10.1080/02643294.2016.1204281