Abstract

Neural oscillations are pivotal to brain function and cognition, but they can be difficult to identify. Researchers engage in careful experimentation to identify their presence, ruling out non-oscillatory processes that could give rise to a similar response. Recently, Doelling and Assaneo have argued against these efforts on the basis that oscillators are heterogeneous, which makes the line to non-oscillators blurred and thereby meaningless to draw. Here, we offer our opposing viewpoint, arguing that we can know whether oscillations are involved, and that we need to know. First, we can know because there are unique properties that only oscillators have, which can be reliably used to identify them – the line is not blurred. These unique properties include eigenfrequency, Arnold Tongue, convergence, and independence. Second, we need to know because there are shared properties, which all oscillators or all oscillators within a subclass have. These shared properties comprise all the information we get once we know there is an oscillator, including neurophysiological, functional, and methodological properties – the fruits of decades of research. We argue that identifying oscillators is crucial for the advancement of research fields as it constrains the possible neural dynamics involved and allows us to make informed predictions on a variety of levels. While neural oscillations are the start and not the end, we have to reach that start.