Taking it all in: Holistic processing of words and faces

As you try to make sense of this sentence, do you find yourself looking at individual letters, or processing each word automatically? If you are an expert English reader, you probably said the latter.

Evidence suggests that at least some word recognition occurs due to holistic processing, or perceptually integrating letters into a unitary whole. The same is true for our recognition of faces – instead of taking in each facial feature individually (e.g., eyes, nose, mouth), we tend to integrate them during perception.

Holistic processing occurs for both words and faces, but does such processing rely on a shared mechanism? This was the question that Paulo Ventura and colleagues (pictured below) set out to answer in a recent paper published in the Psychonomic Society journal Memory & Cognition.

Images of the authors.
Authors of the featured article from left to right: Paulo Ventura, Tina Liu, Francisco Cruz, and Alexandre Pereira.

The Composite Task and prior outcomes supporting holistic processing

To understand what the authors did to answer their question, we first need an understanding of the composite task – a perceptual task commonly used to assess holistic processing of faces. In this task, illustrated in the figure below, participants undergo multiple trials in which they are briefly shown a face (composed of two sides) to study, then shown another composite face at test. At test, participants make same-different judgments – that is, they judge whether the left side of the face matches the left side of the face they saw during study.

Illustration of the composite task.
Abstract illustration of the composite task with left-right face composites, in which participants make same-different judgments for the left side of the face composites.

Prior outcomes using this task support holistic processing of faces. First, consider the congruency effect: The right half of the composite face is irrelevant to making accurate same-different judgments about the left side of the face, but participants generally perform better (i.e., make fewer errors) when the two sides of the faces presented during study are congruent with those presented during test compared to when they are incongruent. For example, on a trial in which a participant should make a same response (“same trial”), they would perform better on trials in which both the left and right sides of the face composites are the same during study and test compared to trials in which the left side of the face is the same at study and test but the right side of the face changes from study to test.

Stroop example.
Fun fact! The classic Stroop effect is another example of a congruency effect in Psychology, in which the word meaning is irrelevant to the main task of identifying the font color, but response time is slower when the two are incongruent compared to when they are congruent.

Second, the congruency effect is drastically reduced when the two parts of the face are misaligned – that is when the right part of the face is presented slightly lower than the left part (as in the image below), producing an alignment by congruency interaction.

Example of a misaligned face composite.
Example of a misaligned face composite.

The congruency effect has also been found when words are used rather than faces (examples are in the table below), suggesting holistic processing of words as well. Does this mean that the mechanisms supporting holistic processing of faces and words are the same? Not necessarily, which is why the authors developed a modified composite task for their experiments.

Examples of words used in the composite task.
Examples of words used in the composite task.

Modified Composite Task

In the modified composite task, composite words were superimposed over composite faces – as in the figure below. Participants were then asked to ignore one stimulus (e.g., the words) while focusing on the other stimulus (e.g., the faces).

Illustration of modified composite task.
Modified Composite Task, which allowed the authors to test their hypothesis by manipulating face and word alignment simultaneously.

This task allowed the authors to investigate whether holistic processing of faces and words interfere with each other and thus share similar mechanisms. If so, the authors predicted that faces would be processed less holistically – as evidenced by a smaller congruency by alignment interaction – when superimposed with aligned rather than misaligned words (because aligned words are also processed holistically).

Ignore words, focus on faces

In the first experiment, undergraduate students from a university in Portugal underwent the modified composite task with instructions to focus on the faces and ignore the words. Participants completed 384 trials that all involved 4 phases: (1) fixation screen, (2) study stimulus (i.e., a face with a 4-letter word overlaid), (3) pattern mask, and (4) test stimulus (i.e., a face with a 4-letter word overlaid). Faces varied in whether they were congruent or incongruent, and aligned or misaligned. The superimposed words could also be aligned or misaligned.

Holistic word processing interferes with holistic face processing

Consistent with the hypothesis that faces and words share mechanisms, the authors observed a significant three-way interaction between face congruency, face alignment, and word alignment on task performance (see figure below). Let’s break that down!

Plot of results.
Performance (sensitivity) on the modified congruency task from Experiment 1 of the featured article. Authors found a face congruency by face alignment interaction when the superimposed words were aligned, but no such interaction when the superimposed words were misaligned.

When the superimposed words were misaligned, an interaction was observed between face congruency and face alignment. Thus, holistic processing of faces was not impacted by irrelevant stimuli (words) when they were misaligned.

Critically, when the superimposed words were aligned, no such face congruency by face alignment interaction was observed!

Thus, whereas holistic face processing was found in the presence of misaligned words, no holistic face processing was found in the presence of aligned words – ultimately suggesting a specific interference with holistic face processing due to word alignment.

Putting it all together

Perceptual processing of faces and words is distinct, with dedicated regions in the brain. But Ventura and colleagues provide initial evidence that holistic processing of faces and words might share cognitive resources.

According to Ventura:

“These findings have important implications for our understanding of a hypothetical shared mechanism (i.e., shared cortical resources and cognitive processes) between face and word processing, and more broadly, the functional architecture of the mind.”

Featured Psychonomic Society article:
Ventura, P., Liu, T.T., Cruz, F., & Pereira, A. (2022). The mechanisms supporting holistic perception of words and faces are not independent. Memory & Cognition. https://doi.org/10.3758/s13421-022-01369-0

Author

  • Michelle Rivers

    Michelle Rivers is a Chancellor's postdoctoral fellow in Psychology at Texas Christian University. Originally from California, they received a B.A. in Psychology from UC Santa Cruz, and completed their M.A. and Ph.D. in Cognitive Psychology at Kent State University working with John Dunlosky. Broadly, their research applies theories of learning and memory to enhance educational practice. Primary aims of their research are to identify, develop, and describe the underlying cognitive mechanisms of techniques that improve self-regulated learning and metacognition.

    View all posts

The Psychonomic Society (Society) is providing information in the Featured Content section of its website as a benefit and service in furtherance of the Society’s nonprofit and tax-exempt status. The Society does not exert editorial control over such materials, and any opinions expressed in the Featured Content articles are solely those of the individual authors and do not necessarily reflect the opinions or policies of the Society. The Society does not guarantee the accuracy of the content contained in the Featured Content portion of the website and specifically disclaims any and all liability for any claims or damages that result from reliance on such content by third parties.

You may also like