I found one here with a bit of googling:
Scroll down to section “4.2. Size and Profile Shape Evolution” and we find this:
4.2. Size and Profile Shape Evolution
One of the most important findings in galaxy evolution studies in the past decade has been the discovery that distant galaxies are more compact than systems of the same mass in the local universe (e.g., Daddi et al. 2005; Trujillo et al. 2007; Buitrago et al. 2008; van Dokkum et al. 2008, 2010; Weinzirl et al. 2011; Baro et al. 2013; Williams et al. 2014).
This change in sizes with time is now well characterized, and the evolution of galaxy sizes at a constant stellar mass selection of M* > 1011 M⊙ can be characterized by a power law of the form R e ~ α (1 + z)β. The value of the power-law slope changes with the galaxy surface brightness type, such that the disk-like galaxies with Sérsic indices n < 2.5 evolve with β = -0.82 ± 0.03, while spheroid-like galaxies with n > 2.5 have β = -1.48 ± 0.04 (Figure 11). This demonstrates that there is a faster evolution in measured sizes for spheroid-like galaxies, which therefore have a more effective increase in size over cosmic time than the disk-like objects.
Figure 11. The average sizes of massive galaxies selected with M* > 1011 M⊙ as imaged in the POWIR (Conselice et al. 2007) z < 2 data and GNS > 1.5 images (Buitrago et al. 2008; Conselice et al. 2011). The size evolution is divided into galaxies with elliptical-like profiles, with Sérsic indices n > 2.5, and disk-like profiles having n < 2.5. The measured effective radius, r e, is plotted as a function of the ratio with the average size of galaxies at the same stellar mass measurements with M* > 1011 M⊙ at z = 0 from Shen et al. (2003).This size evolution is such that the effective radii of massive galaxies increases by up to a factor of five between z = 3 and today at the same stellar mass (e.g., Buitrago et al. 2008; Cassata et al. 2013). The form of this evolution has been investigated to determine whether or not the increase is due to the build up of the entire galaxy or just the inner or outer parts. The data to date show that galaxy growth through sizes is occurring in its outer parts, with the central parts in place at early times (e.g., Carrasco et al. 2010; van Dokkum et al. 2010). This indicates that the build up of massive galaxies is an inside out process, whereby the inner parts of massive galaxies are in place before the outer parts with the same stellar mass density as today (e.g., Hopkins et al. 2009).
An alternative way to investigate this problem is to examine the number of compact and ultra-compact galaxies at various redshifts. There is some controversy over whether or not there exist in the local universe compact galaxies with sizes similar to those seen at high redshifts. However, what is clear is that the number densities of these ultra-compact galaxies declines in relative abundance very steeply at z < 2 (Cassata et al. 2013).
The processes responsible for this increase in sizes at lower redshifts is not well understood, and is currently a source of much debate. The most popular explanation is that this size increase is produced through minor mergers (e.g., Bluck et al. 2012; McLure et al. 2013), although other ideas such as AGN performing work on gas is another idea (e.g., Bluck et al. 2011). However, the outer parts of nearby massive galaxies are too old to have been formed in relatively recent star formation, and the star formation observed at high redshift is not sufficient to produce the observed increase in sizes (Ownsworth et al. 2012).
The major idea for the physical mechanism producing galaxy size evolution is through dry minor mergers, as major mergers are not able to produce the observation of increasing size without significantly increasing mass (e.g., Khochfar & Silk 2006; Naab et al. 2009; Bluck et al. 2012; Oser et al. 2012; Shankar et al. 2013). There is currently some controversy over whether or not the observed minor merger rate is high enough to provide this increase in sizes, with the most massive galaxies with M* > 1011 M⊙ appearing to have enough minor mergers (e.g., Kaviraj et al. 2009) to produce this size evolution (Bluck et al. 2012), but this may not be the case for lower mass systems (e.g., Newman et al. 2012). It does appear however that minor mergers are a significant mechanism for producing low levels of star formation in early-types at z ~ 0.8, as well as for adding significant amount of stellar mass to these galaxies (Kaviraj et al. 2009, 2011). One of the major issues is determining not only the number of minor dry mergers, but also the time-scale for these mergers (Section 3.4) which more simulations would help understand.
Along with the evolution of galaxy sizes, there is also a significant evolution in the underlying structures of galaxies at higher redshifts. One of the cleanest ways to see this is through the evolution of the Sérsic parameter, n (Figure 7). When examining the evolution of derived values of n as a function of redshift for both a stellar mass and at a constant number density selection, it is apparent that galaxies have lower n values at higher redshifts for the same selection (e.g., Buitrago et al. 2013). This has been interpreted by some to imply that these galaxies are more ‘disk-like’ at high redshifts (Bruce et al. 2012), although the morphologies of these systems by visual inspection, and their internal structures and colors, are not similar to modern disks (e.g., Conselice et al. 2011; Mortlock et al. 2013). It appears that these disk-like galaxies, while having light profiles similar to modern disks, are much smaller, have a higher stellar mass, and are often undergoing intense star formation with peculiar morphologies, making them un-disk-like in all other regards. They indeed are likely a type of galaxy with no local counterpart.
Generally that entire webpage explains that in every measurable way, galaxies become increasingly different from how they look in present times, proportional to an increase in redshift. They become smaller in size, have different distributions of shape and structure, etc.
It would be very odd if the JWST suddenly reversed all these trends. It’s ridiculous on it’s face. Will it see weird, unusual, and unexpected things? I’m sure it will. Will they reverse current trends? I’m sure they won’t.