A Method for Investigating Change Blindness in Pigeons (Columba Livia)

This method can help us answer key questions in comparative cognition, such as do pigeons show the same puzzling failures of attention that humans do, and if so, why? The main advantage of this technique is that it can be easily implemented in the standard operant chambers that have been used by behavioral and comparative psychologists for decades. This section described how to train pigeons to peck at stimuli in an operant chamber and accept a reward from the food hopper.

At the beginning of each day's session, weigh the pigeons and place them in the operant chambers. The operant chamber is under computer control, with a temporal resolution and response time of less than one millisecond. Run the pretraining sessions daily, and run 100 trials per session.

For each trial, randomly select a stimulus element and one of the three keys. Then, illuminate the element on the key using a fast LED source for onset with one millisecond and a clear visual presentation. When the pigeon pecks at the key illuminated by the stimulus, set the stimulus to clear and the grain hopper to open for two to three seconds.

Naive and less experienced pigeons may need hand shaping or auto shaping to learn to respond, but experienced pigeons will likely be very quick to peck stimuli. After 100 trials, end the session. Remove the pigeons from the chambers, weigh them, and then return them to their home cages.

Between sessions, adjust the bird's food access time to maintain their running weights to 80 to 85%of their free-feeding weight. Continue the pretraining sessions until the pigeons respond quickly and consistently to all of the experimental stimuli to be used, presented on any of the three keys. For the training sessions, program the stimulus display using the experimental control software.

Several variables must be set. For the ISI in this example, it is set as 250 or zero milliseconds. For the number of stimuli repetitions, have values of one, two, four, eight or 16 be randomly selected for each trial.

Next, for each response key, define an original stimulus display consisting of one or more elements made up of the colors or lines presented during the pretraining sessions. Then, define the change as an addition, deletion or alteration of just one of the display elements on just one key. Next, program the inter-trial intervals to be five seconds with house lights on and all the response keys dark.

A stimulus repetition consists of the following sequence. First, a 250 millisecond presentation of the original display. Second, and ISI of the selected length.

Third, the modified display for 250 milliseconds. Fourth, another ISI of the selected length. During the repetitions, all key pecks are ignored.

Afterwards, immediately illuminate all of the response keys with white light and record the first peck on a response key as the only valid response for the trial. If a bird's response was on the key that displayed a change, provide access to grain from the food hopper for two to three seconds. Otherwise, signal an incorrect response by switching the house lights on and off every half-second for 10 seconds.

Run a total of 100 trials per session and run the sessions daily. Maintain the pigeons at around 85%of their free running weight. Maintain the training regime until the pigeon's responses are stable and better than chance.

Using the flicker paradigm as presented, there was a measurable difference in accuracy with and without an interstimulus interval. Accuracy was significantly reduced when a 250 millisecond ISI was used. With or without the ISI, accuracy significantly increased with the number of repetitions.

In an experiment that manipulated the timing of stimulus displays, there was no change blindness effect with a 30-millisecond presentation and a 15-millisecond ISI. During the same experiment, when presentation and ISI durations were doubled, the change blindness effect returned. There was again a difference in accuracy between trials with and without the interstimulus interval, indicating that display timing influences change blindness.

The idea for this method comes from the flicker task, used to investigate change blindness in humans. The method is precise and translates well into an operant chamber. Generally, individuals new to this method might struggle, because it can take extensive training before birds reach an acceptable level of accuracy.

Be patient. While attempting this procedure, it's important to monitor pigeons'daily progress and be patient. They'll begin with chance accuracy and improve through trial and error learning.

Following this procedure, various aspects of stimulus presentation can be manipulated in order to investigate factors such as the importance of change timing, salience and spatial location. The implications of this technique may extend toward a deeper understanding of attention disorders. The pigeons show us that imperfections of selective attention are not uniquely human and thus may hold some evolutionary significance.