Janina Esins

Alumni of the Department Human Perception, Cognition and Action
Alumni of the Group Recognition and Categorization

Main Focus

Research Group:

Supervisors: and

Prosopagnosia or “face blindness” is a developmental or acquired disorder which destroys the ability to process faces holistically. There is limited understanding of both holistic face processing in general and of the disorder, particularly in its developmental form.

We want to get a deeper insight and understanding of face perception and processing in general as well as its impairment, prosopagnosia. In this study, we cerate an extended test battery to examine face recogntion impairments, the role of the featural and configural face dimensions, dynamic information of faces, etc. These results will then be used to gain a broader understanding of developmental prosopagnosia by examining the impairment of recognition and processing of faces.

Face processing in congenital Prosopagnosia


Recognizing faces is arguably the most important method to identify fellow humans. Prosopagnosia disturbs this ability and therefore severely impairs social interactions. In my projects I focus on the congenital (i.e. inborn) form of prosopagnosia. The manifestations of prosopagnosia are very diverse in terms of severity and affected abilities, such that even face-unrelated impairments (e.g. object recognition deficits) can occur [1].


My goal was to develop a collection of psychophysical tests to assess different aspects of face and object recognition in prosopagnosics and to compare their performance to healthy controls in order to identify and quantify their impairments.


I created 16 psychophysical tests to investigate a large variety of aspects in face and object processing, e.g. recognition of facial expressions, familiar persons, facial motion, objects, navigation, etc [Fig. 1]. We tested a total of 27 prosopagnosics and age-, gender- and education-matched controls. Such a large sample size allowed us to gain new insights about the different manifestations of prosopagnosia.

Fig.1: Examples of stimuli of the different tests. From left to right, first row: Famous face test, Composite face test, Famous choice test. Second row: Sensitivity to features test and Navigation test.


Ten of the sixteen tests revealed significant performance differences between prosopagnosics and controls. For example prosopagnosic exhibited worse identification rate of familiar faces, worse recognition of gender of unfamiliar faces but performed well in navigation tasks and object recognition tasks, etc. Our results show, for example, that prosopagnosics, in comparison to controls, are impaired in processing the spatial distances between the facial features (configuration) [2], [Fig. 2] while they are only slightly impaired in processing the shape of facial features. Another finding is that, in contrast to controls, prosopagnosics do not show any advantage from learning dynamic faces, compared to learning static faces [Fig.3]. This finding is in accordance with a study finding that the pSTS (a brain region known for processing facial and body motion) has a reduced connectivity to the other core regions for face processing in congenital prosopagnosics compared to controls [3].


The large range of individual and general performances of the prosopagnosics across all tests shows that prosopagnosia is even more diverse than expected. In sum, this project shed further light on the impairments caused by congenital prosopagnosia for a better understanding of this condition. It also allows for a better understanding of face processing in general.


[1]  Kress T, and Daum I (2003) Developmental prosopagnosia: a review. Behavioural neurology, 14(3-4), 109-121.

[2]  Esins J, Schultz J, Wallraven, C, Bülthoff I, (2014). Do congenital prosopagnosia and the other-race effect affect the same face recognition mechanisms? Frontiers in Human Neuroscience, 8, 1–14.

[3]  Avidan G, Tanzer M, Hadj-Bouziane F, Liu N, Ungerleider  LG, Behrmann M (2013) Selective dissociation between core and extended regions of the face processing network in congenital prosopagnosia, Cerebral cortex, 24(6), 1565-1578.

Fig.2: Sensitivity of prosopagnosic participants (red) and controls (blue) to featural and configural information of a face. Participants had to rate the similarity (on a Likert-scale from 1 to 7) of two faces either differing only in features or only in configuration. Sensitivity was calculated based on the given ratings. For features the difference between groups was marginally significant (p = .051), for configuration it was significant (p = .001) and there was no significant interaction.

Fig.3: Performance of prosopagnosic participants (red) and controls (blue) for recognizing faces. Faces were both learned and tested either statically or dynamically. While matched controls showed a significantly better performance for dynamic faces than static faces, prosopagnosics did not. Also the interaction of participant group and motion information was significant (p = .03).

Curriculum Vitae


currently PhD student at the Max Planck Institute for Biological Cybernetics, Germany
2003 - 2009 Studies of Biomathematics at the Universität Greifswald, Germany; Grade: German Diplom (equivalent to a Master's degree)
2006 Study abroad at the Massey University, Palmerston North, New Zealand
2002 - 2003 Freiwilliges Ökologisches Jahr (voluntary year of ecological service) at the Institute of Microbiology, Universität Greifswald, Germany
1993 - 2002 Gerhardt-Hauptmann-Gymnasium Wismar, Germany; Abitur (University entrance qualification)


02/2009 - 07/2009

06/2007 - 08/2007

Research assistant / assistant professor at the Institute of Mathematics and Computer Science,  Greifswald, Germany.
Duties: measurement and evaluation of high-frequency EEG data with Matlab02/2009 and 06/2007


08/2009 - 12/2009 Language studies in Peru

Go to Editor View