It’s a lovely Friday evening and you are sitting inside a busy restaurant. Around you is the regular collection of sounds: restaurant goers chatting loudly, the clinking of utensils against plates, glasses hitting tables, and chairs scraping the ground. The server at the table next to you is explaining this evening’s specials. And the music is just a bit too loud.

Despite all this noise, you are able to focus on your conversation with the person sitting across from you. This ability to ignore irrelevant sensory input and focus on more important information is called sensory filtering. Sensory filtering mechanisms allow our brains to process the overwhelming inputs from the environment. For individuals with neurodevelopmental or neuropsychiatric disorders, these sensory filtering abilities can be compromised. For example, the sounds of the restaurant could be too loud and overstimulating.

Suddenly from behind you, a giant crash! One of the overworked servers drops a tray of glasses causing you to startle. Your shoulders scrunch up and your head turns in the direction of the noise. Without realizing it, you engage in a protective behavioural response called the acoustic startle response; this is a fast whole-body flinch to a loud unexpected sound. The acoustic startle response is a commonly used method to test sensory filtering abilities in humans and other species, including rodents and fish. It is especially useful when developing and validating animal models for specific disorders where one of the defining traits is altered sensory filtering. The brain pathways and structures involved in the startle response are found to be highly conserved between species rendering a high translational value.

The amount of response, or the response magnitude, is malleable to change based on other sensory inputs. If a salient sensory stimulus (a prepulse) shortly precedes a startle stimulus, the following startle response is greatly reduced. This reduction in response magnitude is called prepulse inhibition (PPI) and is a measure of sensory filtering. It is assumed that the prepulse elicits an orientating response while other incoming sensory information is suppressed for a short while to not interfere with the orienting response. Individuals with neurodevelopmental or neuropsychiatric disorders, including autistic individuals, often have reduced PPI.

Modifications to Classic Techniques

For decades, startle and PPI have been widely used in pre-clinical and clinical studies to examine pathology-related changes in sensory filtering abilities. One of the benefits of these testing paradigms is the ease and efficiency with which they can be assessed. Classical studies measure PPI at a single startle sound intensity – one that elicits maximum startle responses – combined with various prepulse sound levels. However, it has recently been suggested that this way of testing PPI is insufficient to evaluate the full effect of a prepulse on startle, and that instead a prepulse should be combined with different startle stimulus intensities, including intensities that do not yield maximum startle responses.

The startle response curve is a psychometric input/output function measuring the startle response magnitude (output) in relation to a range of startle sound intensities (input); an individual’s responses can then be fit to a sigmoidal regression function. A prepulse is thought to change this sigmoidal input/output curve in two ways: firstly, a reduced startle reactivity compresses the curve vertically, this is referred to as startle scaling. Secondly, the prepulse changes sensitivity to a startle stimulus in a process called sound scaling which leads to a rightward shift of the input/output curve and changes the startle threshold and the sound levels for half-maximum and maximum startle responses. Startle scaling is thought to be due to changes in the motor portion of the startle response, while sound scaling is presumably due to changes in sound processing. Classical measures of PPI are not able to distinguish between these two components, as only a reduction of startle at one single startle stimulus intensity is observed.

Applying these Modifications

To assess the validity and usefulness of the suggested new method to measure PPI, we tested the Cntnap2 knockout rat – a genetic model with autism-like traits – that has consistently shown PPI disruptions in the past. The new testing paradigm included a range of prepulse and startle sounds. First, we ensured that classical data analysis using the modified testing regime remained consistent with the increased startle and disrupted PPI previously reported in this model.

Using the more advanced analysis, we found that the prepulses resulted in both startle scaling and sound scaling of startle response curves in wildtype rats. In Cntnap2 knockout rats, on the other hand, the prepulses resulted only in startle scaling, while sound scaling was found to be largely impaired, specifically at louder startle sounds. As sound scaling was suggested to be related to sound processing, this suggests that alterations in regions responsible for sound processing are affecting sensory filtering in the Cntnap2 knockout rats.

Broader Implications

Aside from characterizing the Cntnap2 knockout model, this work has much broader implications. Startle and PPI are widely used in neuroscience to measure sensory filtering/sensorimotor gating abilities in models of neurodevelopmental and neuropsychiatric disorders. Also, the ability to rescue altered sensory filtering abilities is crucial for many clinical drug studies. As sensory filtering mechanisms are fundamental building blocks for higher cognitive function, impairments in these abilities are a detriment to the comfort and functionality in the daily lives of individuals. Assessing scaling parameters improves our understanding of the brain mechanisms involved in sensory filtering disruptions and will ultimately benefit the development of therapies.