The coupled pseudostate approximation (CP) introduced by McGovern et al. [Phys. Rev. A 79, 042707 (2009)] to study differential ionization in heavy particle collisions is extended to the case of a Li target. Symmetry properties dependent upon the initial state of the target, and the role of the interaction between the projectile and target nuclei, are discussed. Comparison is made with the recent double differential cross-section measurements (d(2)sigma/dEdq(t)) of LaForge et al. [J. Phys. B 46, 031001 (2013)] for proton impact at 6 MeV and O8+ impact at 1.5 MeV / amu on Li(2s) and Li(2p). It is shown that the proton measurements lie very much in the first Born regime and that the comparison is therefore more of a test of the experiment than of the theory which, at the first Born level, is considered to be quite accurate. By contrast, the O8+ measurements present a nonperturbative scenario which provides a substantive challenge to the full CP theory. Although there are exceptions, the CP approximation generally agrees well with the shape of the O8+ data but there is a drift in normalization between the calculated and measured cross sections with increasing electron ejection energy. This same drift is seen in comparison between CP and measurements of d sigma / dE from Fischer et al. [Phys. Rev. Lett. 109, 113202 (2012)]. The convergence of the CP approximation with respect to the angular momenta of the pseudostates and their number is investigated. It is concluded that the 164-state set employed here should be satisfactory. Comparison is also made with the even more recent fully differential (triple differential) measurements of Hubele et al. [Phys. Rev. Lett. 110, 133201 (2013)] for O8+ impact on Li(2s) and Li(2p) at 1.5 MeV / amu. The CP approximation reproduces the "wings" seen in the Li(2s) measurements, which are due to the interaction between the projectile and target nuclei and the node in the 2s wave function, and the asymmetric shape of the 2p data. The orientational dichroism between the 2(p+ 1) and 2(p-1) states is illustrated. Overall, and while not perfect, the agreement between the pseudstate theory and the experiments is very encouraging.