BPS2026 – The cell as a low-pass filter: Frequency domain analysis of single-cell learning in Stentor coeruleus

One of the features that distinguishes living systems from inanimate matter is their ability to process and store information. The giant ciliate Stentor coeruleus responds to discrete pulses of mechanical stimulation by contracting, which in nature is an escape response to avoid predation. When the cell is presented with repeated stimulation of the same magnitude, it gradually stops responding, representing a primitive form of learning. One way to view this learning process is in terms of signal processing. Thus, the cell “reads” an input in the form of discrete stimulus pulses, and produces an output in the form of a series of contractions. We have previously shown that the output of a Stentor cell depends on the frequency of the input, such that it produces less output on average for higher frequency inputs. In other words, the cell is a low-pass filter. Here, we measure the transfer function and find that it displays a flat response at low-frequencies, a clearly defined corner frequency, and a 20 dB/decade roll off at higher frequencies, thus representing a classical single-pole low-pass filter. By comparing the transfer function to predictions of models, we can now test different models for single-cell learning using a new type of data obtained by frequency domain analysis.