University of Alberta Dictionary of Cognitive Science: Total Turing Test

The Turing test was proposed as a method to determine whether a machine was truly intelligent (Turing, 1950). The core assumption of this test is that verbal behavior is sufficient to determine whether an agent is intelligent enough. The Turing test has been influential for half a century, but many researchers have been concerned about its usefulness. Variations of the original Turing test have been proposed in an attempt to develop a stronger “benchmark” of intelligence.

One example is Harnad’s (1991) Total Turing Test. Harnad (1991) dubs this new and elevated standard the ‘Total Turing Test’. In the Total Turing Test, verbal behaviors are not the sole standard for intelligence: other behaviors are examined too. “‘The candidate must be able to do, in the real world of objects and people, everything that real people can do” (Harnad, 1991, p.44). In other words, the Total Turing Test can only be applied to a robot, or some other agent that is situated and embodied in the physical world.

References: