EIGRP Bounded updates

I was reading about EIGRP in Routing TCP/IP Volume 1 by Jeff Doyle and focusing on the comparisons between it and distance vector and link-state protocols. One characteristic of EIGRP that sets it part from other protocols is that updates are "bounded" meaning that they are only sent to the "affected" neighbors. I was trying to find a way to see this behavior in action so I created this summarization scenario.

R4 is in the middle of the star with R3,R5 and R7 at the edges:

R4-R5 = 192.168.45.0/24
R4-R7 = 192.168.47.0/24
R4-R3 = 192.168.34.0/24

R4 is advertising a summary of 192.168.44.0/22 to R3.
If a new link was brought up in the /22 range, R4 will not send an update to R3.

Here it is in action:

R3#debug eigrp packets update
R4#debug eigrp packets update
R5#debug eigrp packets update

R3#sho ip route eigrp
D 192.168.44.0/22 [90/2681856] via 192.168.34.4, 00:02:54, Serial1/1

R7 is off of R4 serial 1/1, R3 is off of R4 serial 1/0:

R4#sho ip eigrp ne
IP-EIGRP neighbors for process 1
H Address Interface Hold Uptime SRTT RTO Q Seq
(sec) (ms) Cnt Num
1 192.168.47.7 Se1/1 10 00:11:55 75 450 0 8
2 192.168.34.3 Se1/0 13 00:30:39 106 636 0 8
0 192.168.45.5 Se1/2 12 00:32:16 106 636 0 8
R4#

Let's add a new loopback on R5 in the range of the summary:

R5(config)#int lo 1
R5(config-if)#ip address 192.168.46.5 255.255.255.0
R5(config-if)#router eigrp 1
R5(config-router)#network 192.168.46.5 0.0.0.0

*Mar 1 20:02:59.075: EIGRP: Enqueueing UPDATE on Serial1/0 iidbQ un/rely 0/1 serno 8-8
*Mar 1 20:02:59.079: EIGRP: Enqueueing UPDATE on Serial1/0 nbr 192.168.45.4 iidbQ un/rely 0/0 peerQ un/rely 0/0 serno 8-8
*Mar 1 20:02:59.087: EIGRP: Sending UPDATE on Serial1/0 nbr 192.168.45.4
*Mar 1 20:02:59.095: AS 1, Flags 0x0, Seq 8/20 idbQ 0/0 iidbQ un/rely 0/0 peerQ un/rely 0/1 serno 8-8
*Mar 1 20:02:59.235: EIGRP: Received UPDATE on Serial1/0 nbr 192.168.45.4
*Mar 1 20:02:59.235: AS 1, Flags 0x0, Seq 26/8 idbQ 0/0 iidbQ un/rely 0/0 peerQ un/rely 0/0

R5 sends an update to R4. R4 only sends it to R7:

*Mar 1 20:02:59.515: EIGRP: Received UPDATE on Serial1/2 nbr 192.168.45.5
*Mar 1 20:02:59.519: AS 1, Flags 0x0, Seq 8/20 idbQ 0/0 iidbQ un/rely 0/0 peerQ un/rely 0/0
*Mar 1 20:02:59.563: EIGRP: Enqueueing UPDATE on Serial1/1 iidbQ un/rely 0/1 serno 11-11
*Mar 1 20:02:59.567: EIGRP: Enqueueing UPDATE on Serial1/1 nbr 192.168.47.7 iidbQ un/rely 0/0 peerQ un/rely 0/0 serno 11-11
*Mar 1 20:02:59.575: EIGRP: Sending UPDATE on Serial1/1 nbr 192.168.47.7
*Mar 1 20:02:59.579: AS 1, Flags 0x0, Seq 24/8 idbQ 0/0 iidbQ un/rely 0/0 peerQ un/rely 0/1 serno 11-11
*Mar 1 20:02:59.583: EIGRP: Enqueueing UPDATE on Serial1/0 iidbQ un/rely 0/1 serno 11-11
*Mar 1 20:02:59.587: EIGRP: Enqueueing UPDATE on Serial1/0 nbr 192.168.34.3 iidbQ un/rely 0/0 peerQ un/rely 0/0 serno 11-11
*Mar 1 20:02:59.587: EIGRP: Enqueueing UPDATE on Serial1/2 iidbQ un/rely 0/1 serno 11-11
*Mar 1 20:02:59.591: EIGRP: Enqueueing UPDATE on Serial1/2 nbr 192.168.45.5 iidbQ un/rely 0/0 peerQ un/rely 0/0 serno 11-11
*Mar 1 20:02:59.591: EIGRP: Sending UPDATE on Serial1/2 nbr 192.168.45.5
*Mar 1 20:02:59.591: AS 1, Flags 0x0, Seq 26/8 idbQ 0/0 iidbQ un/rely 0/0 peerQ un/rely 0/1 serno 11-11

And of course on R3 we see nothing. What's really interesting is we see R4 "Enqueuing" the update but never actually sending it as it does to R5 and R7.

I am still not sure of one thing though. Is this a fundamental characteristic of EIGRP itself or the fact that we are summarizing? I cannot think of another scenario where this "bounded" update scenario would take place without summarization. If you can, please drop a comment.