We are interested in the internal response of condensed matter. For example, we want to understand how the magnetization inside a material relaxes as a function of time after removing an applied field. A practical application is to understand how magnetically recorded information may erode at long times; in other words how magnetic storage loses its memory. We are particularly interested in how this internal response occurs on the length scale of nanometers, which is technologically important as magneto-electronic devices become increasingly compact. Our fundamental goal is to understand the boundary between small-size thermal effects and quantum behavior. The main experimental techniques that we use are nonresonant spectral hole burning (NSHB) and tickle-field transmission electron microscopy (TEM). In NSHB, a large-field oscillation and a small-field step are utilized to measure the linear response and nonlinear recovery inside a sample. Because the responding degrees of freedom are used as their own local probe, it is the most direct technique available for characterizing the internal response of bulk materials. Tickle-field TEM uses a small oscillating magnetic field to “tickle” the response of a thin-film sample, which is imaged using high-resolution electron microscopy. This technique provides direct ../images of nanoscale response, with unsurpassed sensitivity and resolution. The main theoretical modeling that we use involves small-system thermodynamics, or “nanothermodynamics”. Although originally developed by Terrell Hill more than 40 years ago, we are adapting his ideas and applying them to the internal dynamics of condensed matter. We find that finite-size thermal effects provide a common physical basis for non-Debye relaxation, non-Arrhenius activation, and non-classical critical scaling in a wide variety of materials. A recent development is a thermal distribution for finite systems of interacting particles with spin that we call “Bose-Ising” statistics.