Psychological　distress　affects　college　students＇　academic　performance,　and　attention　allocation　plays　an　important　role　in　learning　process.　In　this　paper,　an　experiment　which　combined　smooth　pursuit　eye　movement　and　alphabets　recognition　tasks　was　introduced,　with　the　aim　of　discovering　differences　in　attention　allocation　between　psychological　distress　students　and　normal　students.　Three　kinds　of　data　were　collected:　the　recording　of　alphabets　recognition　answers,　the　eye　movement　data,　and　the　results　of　psychological　test　scale　K10.　We　did　statistical　analysis　on　right　answers,　and　the　results　of　Analysis　of　Variance(ANOVA)　showed　the　differences　between　psychological　distress　students　and　normal　students　were　not　statistically　significant,　however,　accuracy　changing　with　different　velocities　implied　some　differences.　Then　we　adopted　classification　algorithms,　and　found　the　two　groups　could　be　distinguished　using　eye　movement　features　related　to　attention　allocation,　with　the　highest　accuracy　of　76%.　This　also　indicated　attention　allocation　was　different　between　the　two　groups.