FFA & STS COMBINED The concepts and use of the Fusiform Face Area (FFA) in terms of facial recognition and the Superior Temporal Sulcus (STS) in terms of voice recognition are not new on their own. However, those individual technologies and concepts have evolved on their own and now they are being analysed in terms of how they are perhaps used concurrently when one person does (or tries) to recognize another person. This report will cover what the FFA and STS are in general, prior ideas, frameworks and outcomes that have informed and influenced current research and what the future holds, at least based on current trends for the use of FFA and STS in combination or on their own.

FFA & STS Combined

Subject of Discussion

There is a great amount of debate with the circles that exist in the neuro-psychological field regarding the direct integration, or lack thereof, of the brain regions known as the fusiform facial area (FFA) and the superior temporal sulcus (STS) as they are used to identify a person using speech and/or facial characteristics.

What the FFA & STS Are

The use of a combination of STS and FFA is believed to have first come about through the work of Bruce and Young in 1986. The posed a hypothesis that there was a classical model that could be used to identify a person that would be based on hierarchal and/or distinctive pathy manners for facial and speech perception. This perception could then result in correct identification results (Haxby, Hoffman & Gobbini, 2000).

When it comes to superior temporal sulcus (STS), there is either a top to down approach or vice versa. The former approach to perpetual recognition of a sensory processing stage is established in the higher-level mechanics of STS-driven actions that are constructed by audio-visual cognition like in the alternative models that are sometimes used and mentioned. By contrast, a bottom to top approach begins with perception in different lower-level to rigid higher-level pathways in order to distinguish auditory-only data cognition that is seen only with the more conventional models that exist.

Past Ideas

The prior-mentioned model as crafted and formulated by Bruce and Young in 1986 was fairly shoddy in terms of its accuracy and applications. Indeed, it was overall a poor model as it lacked the explanations for other dissociative impairments relating to existing physical deficit and/or medical injury. There were also concern about the definitive brain regions involved that may or may not be involved. The main concern was that there was not an observable and definitive way to prove ideas and hypotheses one way or another. This does mean that the idea of Bruce and Young were wholly wrong and off-base. It just meant that they could not be proven definitively one or another and thus lacked validity (Haxby, Hoffman & Gobbini, 2000).

Recent & Future Directions

More recent and prospective future directions came to light as the 1900's came to a close in the 1990's. The prior-mentioned problem of not being able to prove ideas related to FTA and STS were eased greatly by rapid advances in technology and the prior-mentioned Haxby was on the forefront of those advances as a result. Haxby, of course, was able to pick up where things left off and Haxby was obviously involved with the prior research so the learning curve was not as steep. Haxby was able to take the model posulated by Bruce and Young and then use neuroimaging techniques to investigate any given function. Looking at the two core structures (the FFA and STS) seemed to render in the form of a single coordinated input that flowed into a single path of processing and analysis for scientists and other analysts (Haxby, Hoffman and Gobbini, 2000).

In terms of the very recent past, there are new methods being introduced that help detect various levels of functional neuroimaging that involve STS and/or FFA (if not both) on one level or another. For example, the use of functional magnetic resonance imaging (fMRI) techniques has been used to measure a single neuron in lesion Macaques primate studies. These techniques and their measurements are similar to human studies that have foxed on face-voice selective areas and this can obviously and completely be related to STS and/or FFA, depending on the application and the situation. Image recognition tasks are being used on conscious subjects to reveal dissociated behaviors when it comes to identifying certain facial characteristics and features such as gender when linkable to things like gender, eye gaze, expression, lip moments when it comes to STS (Schall & von Kriegstein, 2014).

Other alternative models have started to emerge in the modern day above and beyond what has already...

More conventional and more widely accepted models of anatomical localization are useful to help classify the different uni-sensory pathway patterns. However, there are clear and definite clinical limits in using brain lesion injury stidies to establish brain area functions. Schall and his associates investigated FFA and STS as it related to direct communication that occurred even during auditory-only speech. Their study compared and contrasted normal subjects against those whose sufferings were from a perception disorder. For example, the Schall study compared normal subjects to those that were suffering from a perception disorder known as prosopagnosia. This disorder is when a person is unable to recognise any other face (Schall & Von Kriegstein, 2014). The evidence from the study found that there was no significant between the normal subjects and the subjects that had Prosopagnosia in terms of their functional connection to FFA. One recent output of the alternative model is that there is a lengthened multi-sensory view that helped to uncover unconscious perception. There were minute but detectable differences in the facial/auditory areas during auditory-only voice recognition in which an emotional response was clearly also present (Schankin & Wascher, 2007). There is evidence that FFA and STS indirectly communicate in concert and together and this can be measured through the prior-mentioned fMRI technology. The crosstalk between the FFA and STS signatures is noticeable and direct. There is a distinct and functional coupling between FFA and unimodal STS during voice-selected familiar speaker recognition.
Another recent treatise on FFA/STS combinations was found in the work of Iidaka of Nagoya University. He states that the "neuroimaging literature indicates the functional significance of both FFA and STS in face processing; however, the evidence for neural connectivity between the regions is limited, which suggests that these two sites play mutually independent roles in face perception and recognition" (Iidaka, 2014). This is obviously a bit of a hedge between what is being argued in this report and what has been believed prior to the FFA/STS combination framework that existed prior. It would seem there are some that are still not sold on FFA and STS being used together, at least at all times (Iidaka, 2014). Some of the same verbiage can be found in the 2013 work of Pyles et al. (Pyles, Verstynen, Scheider & Tarr, 2013). In a similar fashion, the work of Rhodes et al. (2009) reflects that the STS framework "does not code identity," at least on its own. Indeed, different people are going to have similar (but not identical, obviously) voiceprints (Rhodes, Michie, Hughes & Byatt, 2009). The author of this report found yet another study, published in 2015, that actually combined FFA, STS as well as inferior temporal gyrus (ITG). Indeed, some have taken the FFA/STS combination and taken the concept even further (De Souza, Feitosa, Eifuku, Tamura & Ono, 2008).

Discussion

The alternative model that was aimed at direct speech perception is believe to be controlled prior to recognition of a person as fMRI data also seems to imply that the STS is more complex than the voice-only processing that was done previously. This alternative model manifests and shows how simultaneous information to face/voice sensitive areas thought to cause noisy and uncontrollable environmental inputs actually result with a communicative illusion called the McGurk Effect (Blank, Anwander & Von Kreigstein, 2011).

Tractography 3D modelling systems were used to represent anatomical fibre bundles data visually against observable patterns associated with the FFA and STS areas at each individual level. Diffusion tensor imaging techniques ought differences between auditory and visual connections in magnetic resonance imaging using 2D data analysis that would help identify a person using the general FFA/STS framework. Neuropsychology combined with technological techniques provided evidence for cross-modality between face and voice. This would reveal structural connection patterns which are linked between FFA and voice-sensitive areas in the STS framework that direct voice recognition and thus rejects ideas proposed in the previously discussed models on the subject (Jou et al., 2011). Duly noted are the issues that still persist with tractography methods. This includes connectivity limitations that might induce or cause false-positive pathways (Blank, Anwander & Von Kriegstein, 2011).

There are actually other areas of life and academic study where the subject of FFA/STS comes through and one of those would be the entertainment industry. One such study was conducted in 2014 as it relates to Italian actors and the people that watch them.…