Sentence Verification Task

Background

At their core Sentence Verification Tasks (SVT) are simple cognitive and linguistic assessment tests of people's speed and performance in evaluating statements as 'true' or 'false'.

By manipulating the content of the statements, SVTs can be used to gain an understanding of people's underlying knowledge, reasoning and biases in a variety of domains and for different purposes. For example, SVTs can be used as simple and quick assessments to gauche students' reading comprehension skills. Other researchers use SVTs to study emotional biases such as the 'The Positivity Advantage': People tend to be faster to process statements featuring highly positive words (possibly due to attentional costs of negative information being interpreted as a 'threat').

In 2012, Andrew Shtulman and Joshua Valcarcel designed an SVT as a speedy reasoning task to study the interplay of 'naive' intuitions (e.g. the 'sun moves around the earth') and contradictory scientific knowledge (e.g. 'the earth moves around the sun') that guides people's underlying reasoning about the world. Their results suggest that people's naive and scientific knowledge coexist in people's mind for quite some time.

Millisecond's SVT is based on the procedure published by Shtulman and Valcarcel (2012) but can be edited for use in different domains.

Task Procedure

Participants receive 20 statements in 10 different knowledge domains (Example: Astronomy). Each domain tests 5 concepts (e.g. "planet") with 4 questions per concept (e.g. "Planets are more massive than moons."). Half of the statements (per concept) are 'truth-consistent' for a 'naive' and 'scientific' explanation (e.g. "Planets are more massive than moons"); half of the statements are 'truth-inconsistent' ("Stars are more massive than planets"). Domains are tested in a blocked format, but the order of the domains is determined randomly for each participant. In each trial, participants are instructed to respond to the statements as fast as possible (but no timeout is implemented) by pressing the 'E' button to evaluate the statements as 'true' and the 'I' button to evaluate the statements as 'false'.

What it Measures

A Sentence Verification Task (SVT) measures the speed and accuracy of semantic processing and linguistic comprehension; the implemented Millisecond SVT targets scientific reasoning.

Psychological domains

• Reading Comprehension: Ability to deduce meaning from written text
• Executive Control: Measures the ability to resolve conflict among competing stimuli, involving response inhibition and decision-making
• Scientific Knowledge: The generalized, highly reliable body of understanding about the world that is testable and derived from observation, experimentation, and logical reasoning

Main Performance Metrics

• Overall Accuracy: Proportion correct evaluations (in regard to scientific understanding); main measure of scientific knowledge
• Accuracy Of Inconsistent Truth Statements: Proportion correct evaluations of truth-inconsistent statements; targeted measure of scientific knowledge
• Decision Speed Difference: Difference in response speed to truth-consistent and truth-inconsistent statements; main measure of knowledge coexistence and interference

Psychiatric Conditions

Researchers use SVTs to identify where language or cognitive processing breaks down in specific patient groups, e.g.

• Schizophrenia:
• Parkinson's Disease (PD): Interpreting cross-sectional images like MRIs and X-rays.
• Alzheimer’s Disease (AD)

References

Search Google Scholar for peer-reviewed, published research using the Inquisit Sentence Verification Task.

---

Thomas, E., Hemenway, K., & Restle, Frank. (1981). Processing instructions, markedness, and congruity effects in a sentence verification task. Journal of Experimental Psychology: Human Perception and Performance, 7(3), 701-718.

Hartley, L., Dunne, M., Schwartz, S., & Brown, J. (1986). Effect of noise on cognitive strategies in a sentence verification task. Ergonomics, 29(4), 607-617.

Sanjuán, Forn, Ventura-Campos, Rodríguez-Pujadas, García-Porcar, Belloch, . . . Ávila. (2010). The sentence verification task: A reliable fMRI protocol for mapping receptive language in individual subjects. European Radiology, 20(10), 2432-2438.

Shtulman, A. & Valcarcel, J. (2012). Scientific knowledge suppresses but does not supplant earlier intuitions. Cognition, 124, 209–215.