Table 3.

Statistical testing.

	Groups compared	Test	Null hypothesis	P-value (P<0.05 in bold)
4D, 4G	Mean 2–6 Hz power below −60 mV for attp40 (panel 4D, left side) vs Ca-α1T RNAi (panel 4G, left side)	t-test, two-sided, means of two independent samples	Samples have equal means	8.88×10 −4
4F, 4I	Absolute R-L noduli z score at −120° airflow for attp40 (panel 4F, left side) vs Ca-α1T RNAi (panel 4I, left side)	t-test, two-sided, means of two independent samples	Samples have equal means	1.13×10 −2
4F, 4I	Absolute R/L noduli z score at 120° airflow for attp40 (panel 3F, left side) vs Ca-α1T RNAi (panel 3G, left side)	t-test, two-sided, means of two independent samples	Samples have equal means	1.62×10 −2
6G	Empty-Gal4 vs 41H07-Gal4	t-test, two-sided, means of two independent samples	Samples have equal means	6.21×10 −4
6G	Empty-Gal4 vs VT056655-Gal4	t-test, two-sided, means of two independent samples	Samples have equal means	2.44×10 −2
6L	Empty-Gal4 vs 41H07-Gal4	Watson-Williams test for equality of the means of two or more samples of circular data	Samples have equal means	8.97×10 −7
6L	Empty-Gal4 vs VT056655-Gal4	Watson-Williams test for equality of the means of two or more samples of circular data	Samples have equal means	8.66×10 −7
6Q	Heterozygous (+/del) vs null (del/Δ135)	Watson-Williams test for equality of the means of two or more samples of circular data	Samples have equal means	5.95×10 −9
S1E	No airflow vs airflow	t-test, two-sided, means of two paired samples	Samples have equal means	0.725
S1G	No airflow vs airflow	t-test, two-sided, means of two paired samples	Samples have equal means	0.906
S1H	No airflow vs airflow	t-test, two-sided, means of two paired samples	Samples have equal means	0.290
S4H, S4I	Mean 2–6 Hz power for current injections below - 20 pA for attp40 (panel S4H, left side) vs Ca-α1T RNAi (panel S4I, left side)	t-test, two-sided, means of two independent samples	Samples have equal means	1.91×10 −5