According to Nyquist theory, the most appropriate sampling rate for a HFF setting of 70Hz is:

The Nyquist theorem states that to accurately capture a signal without aliasing, the sampling rate must be at least twice the highest frequency of interest. In the context of EEG signals, when the high-frequency filter (HFF) is set to 70 Hz, the theorem dictates that the minimum sampling rate should be at least twice this frequency, which is 140 Hz.

In practical terms, this means that to properly sample a signal that contains frequencies up to 70 Hz, you need to sample it at a rate that is at least 140 samples per second. This ensures that the signal can be reconstructed accurately without losing information or introducing errors known as aliasing. Thus, selecting a sampling rate of 140 samples per second is the most appropriate choice in this scenario, aligning perfectly with the requirements of the Nyquist theory.