The Cocktail Party Problem defined
The cocktail party is a metaphor for the complex auditory experience that is everyday life. In busy offices, crowded restaurants and family gatherings, a listener is challenged to focus in on signals of interest – such as speech from a particular talker – amid a cacophony of competing conversations and background noise.
In such situations, one’s auditory system is tasked with “auditory scene analysis” (Bergman, 1990). Solving the “cocktail party problem” is to be given the means to single out one voice among many others. It depends on perceptually segregating the different sound sources, selecting one to focus on, and then recognizing and comprehending the message coming from that stream. When we listen to someone in a noisy environment, background noise can mask the signal and create gaps in our perception of the speech. Understanding speech becomes even more complicated when the background noise consists of other meaningful conversations. In such cases, other messages compete and interfere with our understanding of the targeted speaker.
To hear, we use our ears, eyes and brains in a variety of bottom-up and top-down processes. People with normal hearing can understand a conversational in everyday settings quite easily. However, even a minor hearing impairment can disrupt auditory scene analysis. Populations that suffer most from understanding speech in noisy situations are people over 60 and those with hearing loss, whether or not they receive treatment (hearing aids/CIs).
As we age, our auditory abilities decline, especially in the high frequency ranges, due to changes in temporal and spectral coding and in cognitive processing (reduced inhibition, memory and agility). The combination of these factors makes it harder to converse comfortably under noisy conditions, leading to less frequent social interaction.
People with hearing loss have problems understanding speech in noisy environments for two main reasons. First, as fewer sounds reach the ear, the harder it becomes to build a perceptual image of speech in order to separate it from other sounds.
Second, sensory-neural hearing loss also affects the recognition of changes in timing and frequency, elements that are helpful to understanding speech in noise.
In fact, the primary reason for the diagnosis of hearing loss is the inability to follow speech in noisy environments. People with mild or moderate hearing loss report that their inability to differentiate between multiple talkers – or to understand speech in a noisy background – is one of their greatest disabilities (Gatehouse and Noble, 2004).
More recently, the concept of “hidden hearing loss” is gaining ground (Kojawa and Lieberman, 2009). Hidden hearing loss refers to noise-induced damage to the nerve cells that connect the inner ear to the brain, without apparent damage to the hearing organ itself. Hidden hearing loss does not affect our ability to hear even the weakest sounds in quiet environments. It does affect our ability to hear and differentiate sounds under noisy conditions.